Add files via upload

ai.js

@@ -971,6 +971,131 @@ ${voiceCallPrompt}`;
   return data.choices?.[0]?.message?.content || '...';
 }
 
+// 实时语音通话中调用 AI（纯文本输出，不带任何格式标记）
+export async function callRealVoiceAI(contact, userMessage, callMessages = [], initiator = 'user') {
+  // 获取 API 配置
+  let apiUrl, apiKey, apiModel;
+
+  if (contact.useCustomApi) {
+    apiUrl = contact.customApiUrl || '';
+    apiKey = contact.customApiKey || '';
+    apiModel = contact.customModel || '';
+
+    const globalConfig = getApiConfig();
+    if (!apiUrl) apiUrl = globalConfig.url;
+    if (!apiKey) apiKey = globalConfig.key;
+    if (!apiModel) apiModel = globalConfig.model;
+  } else {
+    const globalConfig = getApiConfig();
+    apiUrl = globalConfig.url;
+    apiKey = globalConfig.key;
+    apiModel = globalConfig.model;
+  }
+
+  if (!apiUrl) {
+    throw new Error('请先配置 API 地址');
+  }
+
+  if (!apiModel) {
+    throw new Error('请先选择模型');
+  }
+
+  // 实时语音专用提示词（纯文本，无格式）
+  const realVoicePrompt = `你正在和用户进行实时语音通话。
+
+【重要输出规则】
+- 只输出你说出口的话，不要有任何其他内容
+- 禁止使用小括号描述语气、动作、情绪
+- 禁止使用方括号、尖括号等任何标记
+- 禁止添加旁白、说明、注释
+- 一次输出完整的回复，不需要分段
+
+正确示例：
+喂？在呢在呢，怎么突然打电话过来啦，是不是想我了？
+
+错误示例（禁止）：
+喂？（好奇地）在呢~[开心]
+
+【通话规则】
+- 像真人打电话一样自然交流
+- 符合你的角色设定和性格
+- 积极与用户互动，根据话题自然展开对话
+- 可以说的比较多，像真人聊天`;
+
+  // 构建系统提示词
+  const baseSystemPrompt = buildSystemPrompt(contact, { allowStickers: false, allowMusicShare: false, allowCallRequests: false });
+  const systemPrompt = `${baseSystemPrompt}
+
+【当前场景：实时语音通话中】
+${realVoicePrompt}`;
+
+  // 构建消息
+  const messages = [{ role: 'system', content: systemPrompt }];
+
+  // 添加聊天历史
+  const chatHistory = contact.chatHistory || [];
+  chatHistory.forEach(msg => {
+    if (msg.isRecalled) {
+      messages.push({
+        role: msg.role === 'user' ? 'user' : 'assistant',
+        content: '[用户撤回了一条消息]'
+      });
+      return;
+    }
+    messages.push({
+      role: msg.role === 'user' ? 'user' : 'assistant',
+      content: msg.content
+    });
+  });
+
+  // 添加通话标记
+  if (initiator === 'ai') {
+    messages.push({ role: 'assistant', content: '[你主动拨打了实时语音，用户已接听]' });
+  } else {
+    messages.push({ role: 'user', content: '[用户发起了实时语音，你已接听]' });
+  }
+
+  // 添加通话历史
+  callMessages.forEach(msg => {
+    messages.push({
+      role: msg.role === 'user' ? 'user' : 'assistant',
+      content: msg.content
+    });
+  });
+
+  // 添加当前消息
+  messages.push({ role: 'user', content: userMessage });
+
+  const chatUrl = apiUrl.replace(/\/+$/, '') + '/chat/completions';
+
+  const headers = { 'Content-Type': 'application/json' };
+  if (apiKey) {
+    headers['Authorization'] = `Bearer ${apiKey}`;
+  }
+
+  const response = await fetchWithRetry(
+    chatUrl,
+    {
+      method: 'POST',
+      headers: headers,
+      body: JSON.stringify({
+        model: apiModel,
+        messages: messages,
+        temperature: 1,
+        max_tokens: 500
+      })
+    },
+    { maxRetries: 3 }
+  );
+
+  if (!response.ok) {
+    throw new Error(await formatApiError(response, {}));
+  }
+
+  const data = await response.json();
+  return data.choices?.[0]?.message?.content || '...';
+}
+
 // 视频通话中调用 AI（使用专门的视频通话提示词，包含场景描述）
 // initiator: 'user' 表示用户打给AI，'ai' 表示AI打给用户
 export async function callVideoAI(contact, userMessage, callMessages = [], initiator = 'user') {

audio-storage.js

@@ -0,0 +1,277 @@
+/**
+ * 语音存储模块 - 使用 IndexedDB 存储语音回放
+ */
+
+const DB_NAME = 'WechatVoiceStorage';
+const DB_VERSION = 1;
+const STORE_NAME = 'voiceRecordings';
+
+let db = null;
+
+/**
+ * 初始化数据库
+ */
+export async function initAudioDB() {
+  if (db) return db;
+
+  return new Promise((resolve, reject) => {
+    const request = indexedDB.open(DB_NAME, DB_VERSION);
+
+    request.onerror = () => {
+      console.error('[可乐] IndexedDB 打开失败:', request.error);
+      reject(request.error);
+    };
+
+    request.onsuccess = () => {
+      db = request.result;
+      console.log('[可乐] IndexedDB 初始化成功');
+      resolve(db);
+    };
+
+    request.onupgradeneeded = (event) => {
+      const database = event.target.result;
+
+      // 创建存储对象
+      if (!database.objectStoreNames.contains(STORE_NAME)) {
+        const store = database.createObjectStore(STORE_NAME, { keyPath: 'id', autoIncrement: true });
+        // 索引：按联系人和通话记录查询
+        store.createIndex('contactIndex', 'contactIndex', { unique: false });
+        store.createIndex('callTimestamp', 'callTimestamp', { unique: false });
+        console.log('[可乐] IndexedDB 存储结构创建成功');
+      }
+    };
+  });
+}
+
+/**
+ * 保存语音记录
+ * @param {Object} voiceData - 语音数据
+ * @param {number} voiceData.contactIndex - 联系人索引
+ * @param {number} voiceData.callTimestamp - 通话时间戳
+ * @param {string} voiceData.text - 语音对应的文字
+ * @param {Blob} voiceData.audioBlob - 音频数据
+ * @param {number} voiceData.duration - 时长（秒）
+ * @returns {Promise<number>} 保存的记录 ID
+ */
+export async function saveVoiceRecording(voiceData) {
+  await initAudioDB();
+
+  return new Promise((resolve, reject) => {
+    const transaction = db.transaction([STORE_NAME], 'readwrite');
+    const store = transaction.objectStore(STORE_NAME);
+
+    const record = {
+      contactIndex: voiceData.contactIndex,
+      callTimestamp: voiceData.callTimestamp,
+      text: voiceData.text,
+      audioBlob: voiceData.audioBlob,
+      duration: voiceData.duration,
+      savedAt: Date.now()
+    };
+
+    const request = store.add(record);
+
+    request.onsuccess = () => {
+      console.log('[可乐] 语音保存成功, ID:', request.result);
+      resolve(request.result);
+    };
+
+    request.onerror = () => {
+      console.error('[可乐] 语音保存失败:', request.error);
+      reject(request.error);
+    };
+  });
+}
+
+/**
+ * 批量保存语音记录
+ * @param {Array} voiceDataList - 语音数据数组
+ * @returns {Promise<Array>} 保存的记录 ID 数组
+ */
+export async function saveVoiceRecordings(voiceDataList) {
+  const ids = [];
+  for (const voiceData of voiceDataList) {
+    const id = await saveVoiceRecording(voiceData);
+    ids.push(id);
+  }
+  return ids;
+}
+
+/**
+ * 获取指定通话的所有语音记录
+ * @param {number} contactIndex - 联系人索引
+ * @param {number} callTimestamp - 通话时间戳
+ * @returns {Promise<Array>} 语音记录数组
+ */
+export async function getVoiceRecordingsByCall(contactIndex, callTimestamp) {
+  await initAudioDB();
+
+  return new Promise((resolve, reject) => {
+    const transaction = db.transaction([STORE_NAME], 'readonly');
+    const store = transaction.objectStore(STORE_NAME);
+    const index = store.index('callTimestamp');
+
+    const request = index.getAll(callTimestamp);
+
+    request.onsuccess = () => {
+      // 过滤出指定联系人的记录
+      const records = request.result.filter(r => r.contactIndex === contactIndex);
+      resolve(records);
+    };
+
+    request.onerror = () => {
+      reject(request.error);
+    };
+  });
+}
+
+/**
+ * 获取指定联系人的所有语音记录
+ * @param {number} contactIndex - 联系人索引
+ * @returns {Promise<Array>} 语音记录数组
+ */
+export async function getVoiceRecordingsByContact(contactIndex) {
+  await initAudioDB();
+
+  return new Promise((resolve, reject) => {
+    const transaction = db.transaction([STORE_NAME], 'readonly');
+    const store = transaction.objectStore(STORE_NAME);
+    const index = store.index('contactIndex');
+
+    const request = index.getAll(contactIndex);
+
+    request.onsuccess = () => {
+      resolve(request.result);
+    };
+
+    request.onerror = () => {
+      reject(request.error);
+    };
+  });
+}
+
+/**
+ * 获取单条语音记录
+ * @param {number} id - 记录 ID
+ * @returns {Promise<Object>} 语音记录
+ */
+export async function getVoiceRecording(id) {
+  await initAudioDB();
+
+  return new Promise((resolve, reject) => {
+    const transaction = db.transaction([STORE_NAME], 'readonly');
+    const store = transaction.objectStore(STORE_NAME);
+
+    const request = store.get(id);
+
+    request.onsuccess = () => {
+      resolve(request.result);
+    };
+
+    request.onerror = () => {
+      reject(request.error);
+    };
+  });
+}
+
+/**
+ * 删除语音记录
+ * @param {number} id - 记录 ID
+ */
+export async function deleteVoiceRecording(id) {
+  await initAudioDB();
+
+  return new Promise((resolve, reject) => {
+    const transaction = db.transaction([STORE_NAME], 'readwrite');
+    const store = transaction.objectStore(STORE_NAME);
+
+    const request = store.delete(id);
+
+    request.onsuccess = () => {
+      console.log('[可乐] 语音删除成功, ID:', id);
+      resolve();
+    };
+
+    request.onerror = () => {
+      reject(request.error);
+    };
+  });
+}
+
+/**
+ * 删除指定通话的所有语音记录
+ * @param {number} contactIndex - 联系人索引
+ * @param {number} callTimestamp - 通话时间戳
+ */
+export async function deleteVoiceRecordingsByCall(contactIndex, callTimestamp) {
+  const records = await getVoiceRecordingsByCall(contactIndex, callTimestamp);
+  for (const record of records) {
+    await deleteVoiceRecording(record.id);
+  }
+}
+
+/**
+ * 播放语音记录
+ * @param {number} id - 记录 ID
+ * @returns {Promise<HTMLAudioElement>} 音频元素
+ */
+export async function playVoiceRecording(id) {
+  const record = await getVoiceRecording(id);
+  if (!record || !record.audioBlob) {
+    throw new Error('语音记录不存在');
+  }
+
+  const audioUrl = URL.createObjectURL(record.audioBlob);
+  const audio = new Audio(audioUrl);
+
+  return new Promise((resolve, reject) => {
+    audio.onended = () => {
+      URL.revokeObjectURL(audioUrl);
+      resolve(audio);
+    };
+
+    audio.onerror = (err) => {
+      URL.revokeObjectURL(audioUrl);
+      reject(err);
+    };
+
+    audio.play().catch(reject);
+  });
+}
+
+/**
+ * 获取存储统计信息
+ * @returns {Promise<Object>} 统计信息
+ */
+export async function getStorageStats() {
+  await initAudioDB();
+
+  return new Promise((resolve, reject) => {
+    const transaction = db.transaction([STORE_NAME], 'readonly');
+    const store = transaction.objectStore(STORE_NAME);
+
+    const countRequest = store.count();
+    const allRequest = store.getAll();
+
+    let count = 0;
+    let totalSize = 0;
+
+    countRequest.onsuccess = () => {
+      count = countRequest.result;
+    };
+
+    allRequest.onsuccess = () => {
+      const records = allRequest.result;
+      totalSize = records.reduce((sum, r) => sum + (r.audioBlob?.size || 0), 0);
+      resolve({
+        count,
+        totalSize,
+        totalSizeMB: (totalSize / 1024 / 1024).toFixed(2)
+      });
+    };
+
+    transaction.onerror = () => {
+      reject(transaction.error);
+    };
+  });
+}

chat-func-panel.js

@@ -8,6 +8,7 @@ import { sendMessage, sendPhotoMessage, sendBatchMessages, appendMusicCardMessag
 import { isInGroupChat, sendGroupMessage, sendGroupPhotoMessage, sendGroupBatchMessages, getCurrentGroupIndex, appendGroupMessage, showGroupTypingIndicator, hideGroupTypingIndicator, callGroupAI, enforceGroupChatMemberLimit, appendGroupMusicCardMessage } from './group-chat.js';
 import { startVoiceCall } from './voice-call.js';
 import { startVideoCall } from './video-call.js';
+import { startRealVoiceCall } from './real-voice-call.js';
 import { showMusicPanel, initMusicEvents } from './music.js';
 import { showRedPacketPage } from './red-packet.js';
 import { showTransferPage } from './transfer.js';
@@ -656,6 +657,14 @@ function handleFuncItemClick(func) {
       hideFuncPanel();
       startVideoCall();
       return;
+    case 'realvoice':
+      hideFuncPanel();
+      if (isInGroupChat()) {
+        showToast('群聊暂不支持实时语音', 'info');
+        return;
+      }
+      startRealVoiceCall();
+      return;
     case 'music':
       hideFuncPanel();
       showMusicPanel();

chat.js

@@ -952,6 +952,71 @@ export function renderChatHistory(contact, chatHistory, indexOffset = 0) {
       return;
     }
 
+    // 检查是否是实时语音通话记录消息
+    const realVoiceCallRecordMatch = (msg.content || '').match(/^\[实时语音[：:](.+?)\]$/);
+    if (msg.isRealVoice || realVoiceCallRecordMatch) {
+      const callInfo = realVoiceCallRecordMatch ? realVoiceCallRecordMatch[1] : '00:00';
+      const isDuration = /^\d{2}:\d{2}$/.test(callInfo);
+      const isCancelled = callInfo === '已取消';
+      const isRejected = callInfo === '已拒绝' || callInfo === '对方已拒绝';
+      const isTimeout = callInfo === '对方已取消';
+
+      // 麦克风图标
+      const micIconSVG = `<svg class="wechat-call-record-icon" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round">
+        <path d="M12 1a3 3 0 0 0-3 3v8a3 3 0 0 0 6 0V4a3 3 0 0 0-3-3z"></path>
+        <path d="M19 10v2a7 7 0 0 1-14 0v-2"></path>
+        <line x1="12" y1="19" x2="12" y2="23"></line>
+        <line x1="8" y1="23" x2="16" y2="23"></line>
+      </svg>`;
+
+      let realVoiceCallRecordHTML;
+      if (isDuration) {
+        // 已接通：显示实时语音时长
+        realVoiceCallRecordHTML = `
+          <div class="wechat-call-record wechat-real-voice-record">
+            ${micIconSVG}
+            <span class="wechat-call-record-text">实时语音 ${callInfo}</span>
+          </div>
+        `;
+      } else if (isCancelled) {
+        realVoiceCallRecordHTML = `
+          <div class="wechat-call-record wechat-real-voice-record">
+            ${micIconSVG}
+            <span class="wechat-call-record-text">已取消</span>
+          </div>
+        `;
+      } else if (isRejected) {
+        realVoiceCallRecordHTML = `
+          <div class="wechat-call-record wechat-real-voice-record wechat-call-rejected">
+            ${micIconSVG}
+            <span class="wechat-call-record-text">${callInfo}</span>
+          </div>
+        `;
+      } else if (isTimeout) {
+        realVoiceCallRecordHTML = `
+          <div class="wechat-call-record wechat-real-voice-record">
+            ${micIconSVG}
+            <span class="wechat-call-record-text">对方已取消</span>
+          </div>
+        `;
+      } else {
+        realVoiceCallRecordHTML = `
+          <div class="wechat-call-record wechat-real-voice-record">
+            ${micIconSVG}
+            <span class="wechat-call-record-text">${escapeHtml(callInfo)}</span>
+          </div>
+        `;
+      }
+
+      if (msg.role === 'user') {
+        html += `<div class="wechat-message self" data-msg-index="${index}" data-msg-role="user"><div class="wechat-message-avatar">${getUserAvatarHTML()}</div><div class="wechat-message-content"><div class="wechat-bubble wechat-call-record-bubble">${realVoiceCallRecordHTML}</div></div></div>`;
+      } else {
+        html += `<div class="wechat-message" data-msg-index="${index}" data-msg-role="assistant"><div class="wechat-message-avatar">${avatarContent}</div><div class="wechat-message-content"><div class="wechat-bubble wechat-call-record-bubble">${realVoiceCallRecordHTML}</div></div></div>`;
+      }
+      lastTimestamp = msgTimestamp;
+      return;
+    }
+
     // 检查是否是视频通话记录消息
     const videoCallRecordMatch = (msg.content || '').match(/^\[视频通话[：:](.+?)\]$/);
     if (msg.isVideoCallRecord || videoCallRecordMatch) {
@@ -1534,6 +1599,11 @@ export function appendMessage(role, content, contact, isVoice = false, quote = n
   const messageDiv = document.createElement('div');
   messageDiv.className = `wechat-message ${role === 'user' ? 'self' : ''}`;
 
+  // 计算消息在chatHistory中的索引
+  const msgIndex = contact?.chatHistory ? contact.chatHistory.length - 1 : -1;
+  messageDiv.dataset.msgIndex = msgIndex;
+  messageDiv.dataset.msgRole = role === 'user' ? 'user' : 'assistant';
+
   const firstChar = contact?.name ? contact.name.charAt(0) : '?';
   const avatarContent = role === 'user'
     ? getUserAvatarHTML()

config.js

@@ -120,18 +120,24 @@ export function getMemePromptTemplate() {
   return `##【必须使用】表情包功能
 【重要】你【必须】经常发送表情包！每2-3条回复至少发一个表情包！
 
-使用规则：
-- 格式：<meme>表情名称</meme>
-- 只需要填写表情名称，不需要填写文件ID和扩展名
-- 只能从下面列表选择，不能编造名称
+★★★ 表情包标签格式（必须严格遵守）★★★
+格式：<meme>表情名称</meme>
+- 必须是成对标签：开始标签<meme>和结束标签</meme>缺一不可
+- 表情名称必须从下面列表选择，不能编造
+- 不需要填写文件ID和扩展名，只填表情名称
 
-【绝对禁止 - 最重要的规则！】
+【绝对禁止 - 最重要的规则！违反会导致显示错误！】
 <meme>标签前后【绝对不能】有任何其他文字！必须用 ||| 分隔！
-× 错误：好想你<meme>xxx</meme> ← 绝对禁止！标签和文字混在一起！
-× 错误：<meme>xxx</meme>哈哈 ← 绝对禁止！标签后面有文字！
-× 错误：我很开心<meme>xxx</meme>你呢 ← 绝对禁止！标签夹在文字中间！
-√ 正确：好想你|||<meme>xxx</meme> ← 用|||分开，标签独立！
-√ 正确：<meme>xxx</meme>|||哈哈哈 ← 标签独立一条！
+
+× 致命错误：好想你<meme>xxx</meme> ← 禁止！标签和文字粘在一起！
+× 致命错误：<meme>xxx</meme>哈哈 ← 禁止！标签后面有文字！
+× 致命错误：我很开心<meme>xxx</meme>你呢 ← 禁止！标签夹在文字中间！
+× 致命错误：<meme>xxx ← 禁止！缺少结束标签</meme>！
+× 致命错误：xxx</meme> ← 禁止！缺少开始标签<meme>！
+
+√ 正确写法：好想你|||<meme>小狗摇尾巴</meme> ← 用|||分开！
+√ 正确写法：<meme>喜欢你</meme>|||我真的好喜欢你 ← 标签独立！
+√ 正确写法：哈哈|||<meme>小熊跳舞</meme>|||你太搞笑了
 
 可用表情包列表：
 [
@@ -143,7 +149,7 @@ ${displayNames.join('\n')}
 哈哈哈笑死|||<meme>小熊跳舞</meme>|||你太搞笑了
 <meme>喜欢你</meme>|||我真的好喜欢你
 
-记住：表情包让聊天更生动，【必须】经常使用！但<meme>标签必须独立！`;
+★重要★：<meme>和</meme>必须成对出现！标签必须用|||与文字分开！`;
 }
 
 // 保留旧变量名以兼容，但实际使用时应调用 getMemePromptTemplate()
@@ -227,6 +233,24 @@ export const defaultSettings = {
   groupSelectedModel: '',
   groupModelList: [],
 
+  // ========== 语音功能 API 配置 ==========
+  // STT (语音转文字)
+  sttApiUrl: '',
+  sttApiKey: '',
+  sttModel: '',
+
+  // TTS (文字转语音)
+  ttsApiUrl: '',
+  ttsApiKey: '',
+  ttsModel: '',        // 模型
+  ttsVoice: '',        // 音色
+  ttsSpeed: 1,         // 语速
+  ttsEmotion: '默认',   // 情感
+  ttsProxyUrl: '',     // TTS 代理 URL（用于解决 CORS 问题，如 MiniMax）
+
+  // 实时语音通话开关
+  realVoiceEnabled: true,
+
   // 上下文设置
   contextEnabled: false,
   contextLevel: 5,

main.js

@@ -37,6 +37,8 @@ import { initGroupRedPacket } from './group-red-packet.js';
 import { initGiftEvents } from './gift.js';
 import { initCropper } from './cropper.js';
 import { createFloatingBall, showFloatingBall, hideFloatingBall } from './floating-ball.js';
+import { testSttApi, testTtsApi } from './voice-api.js';
+import { getVoiceRecordingsByContact, deleteVoiceRecording, playVoiceRecording } from './audio-storage.js';
 
 // ========== 历史记录功能 ==========
 let currentHistoryTab = 'listen';
@@ -137,6 +139,12 @@ function renderHistoryContent(contact, tabType) {
     return;
   }
 
+  // 语音回放使用专门的渲染函数
+  if (tabType === 'playback') {
+    renderVoicePlaybackContent(contact);
+    return;
+  }
+
   const context = window.SillyTavern?.getContext?.() || {};
   const userName = context.name1 || '用户';
 
@@ -221,6 +229,130 @@ function renderHistoryContent(contact, tabType) {
   });
 }
 
+// 渲染语音回放内容
+async function renderVoicePlaybackContent(contact) {
+  const contentEl = document.getElementById('wechat-history-content');
+  if (!contentEl) return;
+
+  const contactIndex = currentHistoryContactIndex;
+  if (contactIndex < 0) {
+    contentEl.innerHTML = '<div class="wechat-history-empty"><div class="wechat-history-empty-icon">📭</div><div>请先选择联系人</div></div>';
+    return;
+  }
+
+  // 显示加载状态
+  contentEl.innerHTML = '<div class="wechat-history-empty"><div>加载中...</div></div>';
+
+  try {
+    const recordings = await getVoiceRecordingsByContact(contactIndex);
+
+    if (!recordings || recordings.length === 0) {
+      contentEl.innerHTML = `
+        <div class="wechat-history-empty">
+          <div class="wechat-history-empty-icon" style="color: #07c160;">
+            <svg viewBox="0 0 24 24" width="48" height="48">
+              <path d="M12 1a4 4 0 0 0-4 4v7a4 4 0 0 0 8 0V5a4 4 0 0 0-4-4z" stroke="currentColor" stroke-width="1.5" fill="none"/>
+              <path d="M19 10v2a7 7 0 0 1-14 0v-2M12 19v4M8 23h8" stroke="currentColor" stroke-width="1.5" stroke-linecap="round" fill="none"/>
+            </svg>
+          </div>
+          <div>暂无语音回放记录</div>
+          <div style="font-size: 12px; color: var(--wechat-text-secondary); margin-top: 8px;">实时语音通话结束后可选择保存语音</div>
+        </div>
+      `;
+      return;
+    }
+
+    // 按保存时间倒序排列
+    const sortedRecordings = [...recordings].sort((a, b) => (b.savedAt || 0) - (a.savedAt || 0));
+
+    let html = '<div class="wechat-voice-playback-list">';
+
+    for (const recording of sortedRecordings) {
+      const savedTime = recording.savedAt ? new Date(recording.savedAt).toLocaleString('zh-CN', {
+        month: 'numeric',
+        day: 'numeric',
+        hour: '2-digit',
+        minute: '2-digit'
+      }) : '未知时间';
+
+      const durationSec = Math.round(recording.duration || 0);
+      const durationStr = durationSec > 0 ? `${durationSec}"` : '?秒';
+
+      html += `
+        <div class="wechat-voice-playback-card" data-id="${recording.id}">
+          <div class="wechat-voice-playback-card-header">
+            <span class="wechat-voice-playback-time">${escapeHtml(savedTime)}</span>
+            <div class="wechat-voice-playback-actions">
+              <span class="wechat-voice-playback-duration">${durationStr}</span>
+              <button class="wechat-voice-playback-delete" data-id="${recording.id}" title="删除">×</button>
+            </div>
+          </div>
+          <div class="wechat-voice-playback-content">
+            <div class="wechat-voice-playback-text">${escapeHtml(recording.text || '')}</div>
+            <button class="wechat-voice-playback-btn" data-id="${recording.id}" title="播放">
+              <svg viewBox="0 0 24 24" width="20" height="20"><polygon points="5,3 19,12 5,21" fill="currentColor"/></svg>
+            </button>
+          </div>
+        </div>
+      `;
+    }
+
+    html += '</div>';
+    contentEl.innerHTML = html;
+
+    // 绑定播放按钮事件
+    contentEl.querySelectorAll('.wechat-voice-playback-btn').forEach(btn => {
+      btn.addEventListener('click', async (e) => {
+        e.stopPropagation();
+        const id = parseInt(btn.dataset.id);
+
+        try {
+          btn.disabled = true;
+          btn.innerHTML = '<svg viewBox="0 0 24 24" width="20" height="20"><rect x="6" y="4" width="4" height="16" fill="currentColor"/><rect x="14" y="4" width="4" height="16" fill="currentColor"/></svg>';
+
+          await playVoiceRecording(id);
+        } catch (err) {
+          console.error('[可乐] 播放语音失败:', err);
+          showToast('播放失败', '⚠️');
+        } finally {
+          btn.disabled = false;
+          btn.innerHTML = '<svg viewBox="0 0 24 24" width="20" height="20"><polygon points="5,3 19,12 5,21" fill="currentColor"/></svg>';
+        }
+      });
+    });
+
+    // 绑定删除按钮事件
+    contentEl.querySelectorAll('.wechat-voice-playback-delete').forEach(btn => {
+      btn.addEventListener('click', async (e) => {
+        e.stopPropagation();
+        const id = parseInt(btn.dataset.id);
+
+        if (confirm('确定要删除这条语音吗？')) {
+          try {
+            await deleteVoiceRecording(id);
+            showToast('已删除', '✓');
+            // 重新渲染
+            renderVoicePlaybackContent(contact);
+          } catch (err) {
+            console.error('[可乐] 删除语音失败:', err);
+            showToast('删除失败', '⚠️');
+          }
+        }
+      });
+    });
+
+  } catch (err) {
+    console.error('[可乐] 加载语音记录失败:', err);
+    contentEl.innerHTML = `
+      <div class="wechat-history-empty">
+        <div class="wechat-history-empty-icon">⚠️</div>
+        <div>加载失败</div>
+        <div style="font-size: 12px; color: var(--wechat-text-secondary);">${escapeHtml(err.message || '')}</div>
+      </div>
+    `;
+  }
+}
+
 function escapeHtml(text) {
   if (!text) return '';
   const div = document.createElement('div');
@@ -1679,6 +1811,13 @@ function bindEvents() {
         return;
       }
 
+      if (service === 'voice-api') {
+        allPanels.filter(p => p !== 'wechat-voice-api-panel').forEach(p => document.getElementById(p)?.classList.add('hidden'));
+        const panel = document.getElementById('wechat-voice-api-panel');
+        panel?.classList.toggle('hidden');
+        return;
+      }
+
       const label = item.querySelector('span')?.textContent || '该';
       showToast(`"${label}" 功能开发中...`, 'info');
     });
@@ -2142,6 +2281,140 @@ function bindEvents() {
     }
   });
 
+  // ===== 语音 API 面板事件 =====
+  // 关闭按钮
+  document.getElementById('wechat-voice-api-close')?.addEventListener('click', () => {
+    document.getElementById('wechat-voice-api-panel')?.classList.add('hidden');
+  });
+
+  // STT 密钥可见性切换
+  document.getElementById('wechat-stt-key-toggle')?.addEventListener('click', () => {
+    const keyInput = document.getElementById('wechat-stt-api-key');
+    if (keyInput) {
+      keyInput.type = keyInput.type === 'password' ? 'text' : 'password';
+    }
+  });
+
+  // TTS 密钥可见性切换
+  document.getElementById('wechat-tts-key-toggle')?.addEventListener('click', () => {
+    const keyInput = document.getElementById('wechat-tts-api-key');
+    if (keyInput) {
+      keyInput.type = keyInput.type === 'password' ? 'text' : 'password';
+    }
+  });
+
+  // 测试 STT API
+  document.getElementById('wechat-voice-api-test-stt')?.addEventListener('click', async () => {
+    const btn = document.getElementById('wechat-voice-api-test-stt');
+    const originalText = btn?.textContent;
+    if (btn) {
+      btn.textContent = '测试中...';
+      btn.disabled = true;
+    }
+
+    try {
+      // 先保存当前配置
+      const settings = getSettings();
+      settings.sttApiUrl = document.getElementById('wechat-stt-api-url')?.value?.trim() || '';
+      settings.sttApiKey = document.getElementById('wechat-stt-api-key')?.value?.trim() || '';
+      settings.sttModel = document.getElementById('wechat-stt-model')?.value?.trim() || '';
+
+      await testSttApi();
+      showToast('STT 连接成功！', '✓');
+    } catch (err) {
+      console.error('[可乐] STT 测试失败:', err);
+      showToast('STT 测试失败: ' + err.message, '⚠️');
+    } finally {
+      if (btn) {
+        btn.textContent = originalText;
+        btn.disabled = false;
+      }
+    }
+  });
+
+  // 测试 TTS API
+  document.getElementById('wechat-voice-api-test-tts')?.addEventListener('click', async () => {
+    const btn = document.getElementById('wechat-voice-api-test-tts');
+    const originalText = btn?.textContent;
+    if (btn) {
+      btn.textContent = '测试中...';
+      btn.disabled = true;
+    }
+
+    try {
+      // 先保存当前配置
+      const settings = getSettings();
+      settings.ttsApiUrl = document.getElementById('wechat-tts-api-url')?.value?.trim() || '';
+      settings.ttsApiKey = document.getElementById('wechat-tts-api-key')?.value?.trim() || '';
+      settings.ttsModel = document.getElementById('wechat-tts-model')?.value?.trim() || '';
+      settings.ttsVoice = document.getElementById('wechat-tts-voice')?.value?.trim() || '';
+      settings.ttsSpeed = parseFloat(document.getElementById('wechat-tts-speed')?.value) || 1;
+      settings.ttsEmotion = document.getElementById('wechat-tts-emotion')?.value?.trim() || '默认';
+      settings.ttsProxyUrl = document.getElementById('wechat-tts-proxy-url')?.value?.trim() || '';
+
+      const audioBlob = await testTtsApi();
+      console.log('[可乐] TTS 测试返回音频:', {
+        size: audioBlob?.size,
+        type: audioBlob?.type
+      });
+
+      if (!audioBlob || audioBlob.size < 100) {
+        throw new Error('返回的音频数据无效');
+      }
+
+      // 播放测试音频
+      const audioUrl = URL.createObjectURL(audioBlob);
+      const audio = new Audio(audioUrl);
+      audio.volume = 1.0;
+
+      await new Promise((resolve, reject) => {
+        audio.onended = () => {
+          URL.revokeObjectURL(audioUrl);
+          resolve();
+        };
+        audio.onerror = (e) => {
+          URL.revokeObjectURL(audioUrl);
+          reject(new Error('音频播放失败'));
+        };
+        audio.play().then(() => {
+          console.log('[可乐] 测试音频开始播放');
+        }).catch(reject);
+      });
+
+      showToast('TTS 测试成功！正在播放', '✓');
+    } catch (err) {
+      console.error('[可乐] TTS 测试失败:', err);
+      showToast('TTS 测试失败: ' + err.message, '⚠️');
+    } finally {
+      if (btn) {
+        btn.textContent = originalText;
+        btn.disabled = false;
+      }
+    }
+  });
+
+  // 保存语音 API 配置
+  document.getElementById('wechat-voice-api-save')?.addEventListener('click', () => {
+    const settings = getSettings();
+
+    // STT 配置
+    settings.sttApiUrl = document.getElementById('wechat-stt-api-url')?.value?.trim() || '';
+    settings.sttApiKey = document.getElementById('wechat-stt-api-key')?.value?.trim() || '';
+    settings.sttModel = document.getElementById('wechat-stt-model')?.value?.trim() || '';
+
+    // TTS 配置
+    settings.ttsApiUrl = document.getElementById('wechat-tts-api-url')?.value?.trim() || '';
+    settings.ttsApiKey = document.getElementById('wechat-tts-api-key')?.value?.trim() || '';
+    settings.ttsModel = document.getElementById('wechat-tts-model')?.value?.trim() || '';
+    settings.ttsVoice = document.getElementById('wechat-tts-voice')?.value?.trim() || '';
+    settings.ttsSpeed = parseFloat(document.getElementById('wechat-tts-speed')?.value) || 1;
+    settings.ttsEmotion = document.getElementById('wechat-tts-emotion')?.value?.trim() || '默认';
+    settings.ttsProxyUrl = document.getElementById('wechat-tts-proxy-url')?.value?.trim() || '';
+
+    requestSave();
+    showToast('语音 API 配置已保存', '✓');
+  });
+
   // 绑定联系人点击
   refreshContactsList();
 }

message-menu.js

@@ -688,6 +688,14 @@ export function bindMessageBubbleEvents(container) {
 
 // 获取真实的消息索引（排除时间标签等）
 function getRealMsgIndex(container, msgElement) {
+  // 优先从元素属性获取（新消息会有这个属性）
+  if (msgElement?.dataset?.msgIndex !== undefined) {
+    const idx = parseInt(msgElement.dataset.msgIndex);
+    if (!isNaN(idx) && idx >= 0) {
+      return idx;
+    }
+  }
+
   const settings = getSettings();
   const contact = settings.contacts[currentChatIndex];
   if (!contact || !contact.chatHistory) return -1;
@@ -699,7 +707,7 @@ function getRealMsgIndex(container, msgElement) {
   if (visualIndex < 0) return -1;
 
   // 需要计算真实索引（chatHistory中可能包含marker消息和撤回消息）
-  // 注意：包含 ||| 的消息在渲染时会被拆分成多条可视消息，需要正确计算
+  // 注意：包含 ||| 或 <meme> 的消息在渲染时会被拆分成多条可视消息，需要正确计算
   let realIndex = -1;
   let visualCount = 0;
 
@@ -712,9 +720,10 @@ function getRealMsgIndex(container, msgElement) {
     let visualMsgCount = 1;
     const content = msg.content || '';
     const isSpecial = msg.isVoice || msg.isSticker || msg.isPhoto || msg.isMusic;
-    if (!isSpecial && content.indexOf('|||') >= 0) {
-      // 按 ||| 分割后有多少个非空部分
-      const parts = content.split('|||').map(p => p.trim()).filter(p => p);
+    // 检查是否包含 ||| 或 <meme> 标签（这些会导致消息被分割显示）
+    if (!isSpecial && (content.indexOf('|||') >= 0 || /<\s*meme\s*>/i.test(content))) {
+      // 使用 splitAIMessages 计算实际分割数量
+      const parts = splitAIMessages(content).filter(p => p && p.trim());
       visualMsgCount = parts.length || 1;
     }
 

phone-html.js

@@ -228,15 +228,16 @@ export function generatePhoneHTML() {
                 <div class="wechat-func-item" data-func="photo"><div class="wechat-func-icon"><svg viewBox="0 0 24 24"><rect x="3" y="3" width="18" height="18" rx="2" stroke="currentColor" stroke-width="1.5" fill="none"/><circle cx="8.5" cy="8.5" r="1.5" fill="currentColor"/><path d="M21 15l-5-5L5 21" stroke="currentColor" stroke-width="1.5" fill="none"/></svg></div><span>照片</span></div>
                 <div class="wechat-func-item" data-func="voicecall"><div class="wechat-func-icon"><svg viewBox="0 0 24 24"><path d="M5 4h4l2 5-2.5 1.5a11 11 0 005 5L15 13l5 2v4a2 2 0 01-2 2A16 16 0 013 6a2 2 0 012-2" stroke="currentColor" stroke-width="1.5" fill="none" stroke-linecap="round" stroke-linejoin="round"/></svg></div><span>语音通话</span></div>
                 <div class="wechat-func-item" data-func="videocall"><div class="wechat-func-icon"><svg viewBox="0 0 24 24"><rect x="2" y="6" width="13" height="12" rx="2" stroke="currentColor" stroke-width="1.5" fill="none"/><path d="M22 8l-7 4 7 4V8z" stroke="currentColor" stroke-width="1.5" fill="none"/></svg></div><span>视频通话</span></div>
+                <div class="wechat-func-item" data-func="realvoice"><div class="wechat-func-icon" style="background: linear-gradient(135deg, #00bcd4, #009688);"><svg viewBox="0 0 24 24"><path d="M12 1a4 4 0 00-4 4v7a4 4 0 008 0V5a4 4 0 00-4-4z" stroke="currentColor" stroke-width="1.5" fill="none"/><path d="M19 10v2a7 7 0 01-14 0v-2M12 19v4M8 23h8" stroke="currentColor" stroke-width="1.5" stroke-linecap="round" fill="none"/></svg></div><span>实时语音</span></div>
                 <div class="wechat-func-item" data-func="location"><div class="wechat-func-icon"><svg viewBox="0 0 24 24"><path d="M12 2C8.13 2 5 5.13 5 9c0 5.25 7 13 7 13s7-7.75 7-13c0-3.87-3.13-7-7-7z" stroke="currentColor" stroke-width="1.5" fill="none"/><circle cx="12" cy="9" r="2.5" fill="currentColor"/></svg></div><span>位置</span></div>
                 <div class="wechat-func-item" data-func="redpacket"><div class="wechat-func-icon"><svg viewBox="0 0 24 24"><rect x="4" y="2" width="16" height="20" rx="2" stroke="currentColor" stroke-width="1.5" fill="none"/><circle cx="12" cy="10" r="3" stroke="currentColor" stroke-width="1.5" fill="none"/><path d="M4 8h16" stroke="currentColor" stroke-width="1.5"/></svg></div><span>红包</span></div>
                 <div class="wechat-func-item" data-func="gift"><div class="wechat-func-icon"><svg viewBox="0 0 24 24"><rect x="3" y="8" width="18" height="13" rx="2" stroke="currentColor" stroke-width="1.5" fill="none"/><path d="M12 8v13M3 12h18" stroke="currentColor" stroke-width="1.5"/><path d="M12 8c-2-4-6-4-6 0s4 0 6 0c2 0 6-4 6 0s-4 4-6 0" stroke="currentColor" stroke-width="1.5" fill="none"/></svg></div><span>礼物</span></div>
                 <div class="wechat-func-item" data-func="transfer"><div class="wechat-func-icon"><svg viewBox="0 0 24 24"><rect x="2" y="4" width="20" height="16" rx="2" stroke="currentColor" stroke-width="1.5" fill="none"/><path d="M2 10h20" stroke="currentColor" stroke-width="1.5"/><path d="M6 15h4M14 15h4" stroke="currentColor" stroke-width="1.5" stroke-linecap="round"/></svg></div><span>转账</span></div>
-                <div class="wechat-func-item" data-func="multi"><div class="wechat-func-icon"><svg viewBox="0 0 24 24"><path d="M21 15a2 2 0 01-2 2H7l-4 4V5a2 2 0 012-2h14a2 2 0 012 2z" stroke="currentColor" stroke-width="1.5" fill="none"/><path d="M8 9h8M8 13h5" stroke="currentColor" stroke-width="1.5" stroke-linecap="round"/></svg></div><span>多条消息</span></div>
               </div>
             </div>
             <div class="wechat-func-page" data-page="1">
               <div class="wechat-func-grid">
+                <div class="wechat-func-item" data-func="multi"><div class="wechat-func-icon"><svg viewBox="0 0 24 24"><path d="M21 15a2 2 0 01-2 2H7l-4 4V5a2 2 0 012-2h14a2 2 0 012 2z" stroke="currentColor" stroke-width="1.5" fill="none"/><path d="M8 9h8M8 13h5" stroke="currentColor" stroke-width="1.5" stroke-linecap="round"/></svg></div><span>多条消息</span></div>
                 <div class="wechat-func-item" data-func="voice"><div class="wechat-func-icon"><svg viewBox="0 0 24 24"><path d="M12 1a4 4 0 00-4 4v7a4 4 0 008 0V5a4 4 0 00-4-4z" stroke="currentColor" stroke-width="1.5" fill="none"/><path d="M19 10v2a7 7 0 01-14 0v-2M12 19v4M8 23h8" stroke="currentColor" stroke-width="1.5" stroke-linecap="round" fill="none"/></svg></div><span>语音输入</span></div>
                 <div class="wechat-func-item" data-func="listen"><div class="wechat-func-icon"><svg viewBox="0 0 24 24"><path d="M3 18v-6a9 9 0 0118 0v6" stroke="currentColor" stroke-width="1.5" fill="none"/><path d="M21 19a2 2 0 01-2 2h-1a2 2 0 01-2-2v-3a2 2 0 012-2h3v5z" stroke="currentColor" stroke-width="1.5" fill="none"/><path d="M3 19a2 2 0 002 2h1a2 2 0 002-2v-3a2 2 0 00-2-2H3v5z" stroke="currentColor" stroke-width="1.5" fill="none"/></svg></div><span>一起听</span></div>
                 <div class="wechat-func-item" data-func="music"><div class="wechat-func-icon"><svg viewBox="0 0 24 24"><path d="M9 18V5l12-2v13" stroke="currentColor" stroke-width="1.5" fill="none" stroke-linecap="round" stroke-linejoin="round"/><circle cx="6" cy="18" r="3" stroke="currentColor" stroke-width="1.5" fill="none"/><circle cx="18" cy="16" r="3" stroke="currentColor" stroke-width="1.5" fill="none"/></svg></div><span>音乐</span></div>
@@ -292,6 +293,7 @@ export function generatePhoneHTML() {
       ${generateModalsHTML(settings)}
       ${generateVoiceCallPageHTML()}
       ${generateVideoCallPageHTML()}
+      ${generateRealVoiceCallPageHTML()}
       ${generateMusicPanelHTML()}
       ${generateListenTogetherHTML()}
       ${generateMomentsPageHTML()}
@@ -767,6 +769,7 @@ function generateServicePageHTML(settings) {
           <div class="wechat-service-section-title">AI功能</div>
           <div class="wechat-service-grid">
             <div class="wechat-service-item" data-service="meme-stickers"><div class="wechat-service-icon purple" style="background: linear-gradient(135deg, #9c27b0, #e91e63);"><svg viewBox="0 0 24 24"><rect x="3" y="3" width="18" height="18" rx="2" stroke="currentColor" stroke-width="1.5" fill="none"/><circle cx="9" cy="9" r="1.5" fill="currentColor"/><circle cx="15" cy="9" r="1.5" fill="currentColor"/><path d="M7 14c1.5 3 4 4 5 4s3.5-1 5-4" stroke="currentColor" stroke-width="1.5" stroke-linecap="round" fill="none"/></svg></div><span>Meme表情</span></div>
+            <div class="wechat-service-item" data-service="voice-api"><div class="wechat-service-icon" style="background: linear-gradient(135deg, #00bcd4, #009688);"><svg viewBox="0 0 24 24"><path d="M12 1a4 4 0 00-4 4v7a4 4 0 008 0V5a4 4 0 00-4-4z" stroke="currentColor" stroke-width="1.5" fill="none"/><path d="M19 10v2a7 7 0 01-14 0v-2M12 19v4M8 23h8" stroke="currentColor" stroke-width="1.5" stroke-linecap="round" fill="none"/></svg></div><span>语音API</span></div>
           </div>
         </div>
         <div class="wechat-service-section">
@@ -802,6 +805,77 @@ function generateServicePageHTML(settings) {
             </div>
           </div>
         </div>
+        <!-- 语音 API 设置面板 -->
+        <div class="wechat-service-panel hidden" id="wechat-voice-api-panel">
+          <div class="wechat-panel-header">
+            <span class="wechat-panel-title">语音 API 设置</span>
+            <button class="wechat-panel-close" data-panel="wechat-voice-api-panel">×</button>
+          </div>
+          <div style="padding: 16px; max-height: 70vh; overflow-y: auto;">
+            <div style="font-size: 13px; font-weight: bold; color: #00bcd4; margin-bottom: 10px;">语音识别 (STT)</div>
+            <div style="font-size: 11px; color: var(--wechat-text-secondary); margin-bottom: 8px;">将语音转换为文字</div>
+            <div style="margin-bottom: 12px;">
+              <div style="font-size: 12px; margin-bottom: 4px;">API 地址</div>
+              <input type="text" class="wechat-settings-input" id="wechat-stt-api-url" placeholder="https://api.example.com/v1/audio/transcriptions" value="${settings.sttApiUrl || ''}" style="width: 100%; box-sizing: border-box;">
+            </div>
+            <div style="margin-bottom: 12px;">
+              <div style="font-size: 12px; margin-bottom: 4px;">API 密钥</div>
+              <div class="wechat-settings-input-wrapper">
+                <input type="password" class="wechat-settings-input" id="wechat-stt-api-key" placeholder="sk-..." value="${settings.sttApiKey || ''}" style="width: 100%; box-sizing: border-box;">
+                <button class="wechat-settings-eye-btn" id="wechat-stt-key-toggle"><svg width="18" height="18" viewBox="0 0 24 24" fill="none"><path d="M1 12s4-8 11-8 11 8 11 8-4 8-11 8-11-8-11-8z" stroke="currentColor" stroke-width="2"/><circle cx="12" cy="12" r="3" stroke="currentColor" stroke-width="2"/></svg></button>
+              </div>
+            </div>
+            <div style="margin-bottom: 16px;">
+              <div style="font-size: 12px; margin-bottom: 4px;">模型</div>
+              <input type="text" class="wechat-settings-input" id="wechat-stt-model" placeholder="whisper-1 或 iic/SenseVoiceSmall" value="${settings.sttModel || ''}" style="width: 100%; box-sizing: border-box;">
+            </div>
+
+            <div style="border-top: 1px solid var(--wechat-border); margin: 16px 0;"></div>
+
+            <div style="font-size: 13px; font-weight: bold; color: #009688; margin-bottom: 10px;">语音合成 (TTS)</div>
+            <div style="font-size: 11px; color: var(--wechat-text-secondary); margin-bottom: 8px;">将文字转换为语音</div>
+            <div style="margin-bottom: 12px;">
+              <div style="font-size: 12px; margin-bottom: 4px;">API 地址</div>
+              <input type="text" class="wechat-settings-input" id="wechat-tts-api-url" placeholder="https://api.example.com/v1/audio/speech" value="${settings.ttsApiUrl || ''}" style="width: 100%; box-sizing: border-box;">
+            </div>
+            <div style="margin-bottom: 12px;">
+              <div style="font-size: 12px; margin-bottom: 4px;">API 密钥</div>
+              <div class="wechat-settings-input-wrapper">
+                <input type="password" class="wechat-settings-input" id="wechat-tts-api-key" placeholder="sk-..." value="${settings.ttsApiKey || ''}" style="width: 100%; box-sizing: border-box;">
+                <button class="wechat-settings-eye-btn" id="wechat-tts-key-toggle"><svg width="18" height="18" viewBox="0 0 24 24" fill="none"><path d="M1 12s4-8 11-8 11 8 11 8-4 8-11 8-11-8-11-8z" stroke="currentColor" stroke-width="2"/><circle cx="12" cy="12" r="3" stroke="currentColor" stroke-width="2"/></svg></button>
+              </div>
+            </div>
+            <div style="margin-bottom: 12px;">
+              <div style="font-size: 12px; margin-bottom: 4px;">模型</div>
+              <input type="text" class="wechat-settings-input" id="wechat-tts-model" placeholder="gemini-2.5-flash-preview-tts / tts-1" value="${settings.ttsModel || ''}" style="width: 100%; box-sizing: border-box;">
+            </div>
+            <div style="margin-bottom: 12px;">
+              <div style="font-size: 12px; margin-bottom: 4px;">音色</div>
+              <input type="text" class="wechat-settings-input" id="wechat-tts-voice" placeholder="achird / alloy / nova" value="${settings.ttsVoice || ''}" style="width: 100%; box-sizing: border-box;">
+            </div>
+            <div style="display: flex; gap: 10px; margin-bottom: 12px;">
+              <div style="flex: 1;">
+                <div style="font-size: 12px; margin-bottom: 4px;">语速</div>
+                <input type="number" class="wechat-settings-input" id="wechat-tts-speed" placeholder="1.0" value="${settings.ttsSpeed || 1}" min="0.5" max="2" step="0.1" style="width: 100%; box-sizing: border-box;">
+              </div>
+              <div style="flex: 1;">
+                <div style="font-size: 12px; margin-bottom: 4px;">情感</div>
+                <input type="text" class="wechat-settings-input" id="wechat-tts-emotion" placeholder="默认" value="${settings.ttsEmotion || '默认'}" style="width: 100%; box-sizing: border-box;">
+              </div>
+            </div>
+            <div style="margin-bottom: 12px;">
+              <div style="font-size: 12px; margin-bottom: 4px;">代理 URL <span style="color: #999; font-weight: normal;">(MiniMax 需要)</span></div>
+              <input type="text" class="wechat-settings-input" id="wechat-tts-proxy-url" placeholder="http://你的服务器:3001" value="${settings.ttsProxyUrl || ''}" style="width: 100%; box-sizing: border-box;">
+            </div>
+
+            <div style="display: flex; gap: 10px; margin-top: 16px;">
+              <button class="wechat-btn wechat-btn-small" id="wechat-voice-api-test-stt" style="flex: 1; background: #00bcd4; color: white;">测试 STT</button>
+              <button class="wechat-btn wechat-btn-small" id="wechat-voice-api-test-tts" style="flex: 1; background: #009688; color: white;">测试 TTS</button>
+            </div>
+            <button class="wechat-btn wechat-btn-primary wechat-btn-block" id="wechat-voice-api-save" style="margin-top: 10px;">保存配置</button>
+            <div id="wechat-voice-api-status" style="font-size: 12px; color: var(--wechat-text-secondary); margin-top: 8px; text-align: center;"></div>
+          </div>
+        </div>
       </div>
     </div>
   `;
@@ -944,6 +1018,23 @@ function generateModalsHTML(settings) {
         </div>
       </div>
     </div>
+    <!-- 语音回放选择弹窗 -->
+    <div id="wechat-voice-save-modal" class="wechat-modal hidden">
+      <div class="wechat-modal-content" style="position: relative; max-width: 400px;">
+        <button class="wechat-modal-close-x" id="wechat-voice-save-cancel" title="关闭">×</button>
+        <div class="wechat-modal-title">保存语音回放</div>
+        <div class="wechat-voice-save-hint" style="font-size: 12px; color: var(--wechat-text-secondary); margin-bottom: 12px;">
+          选择想保留的语音，以后可以在聊天记录中回放
+        </div>
+        <div class="wechat-voice-save-list" id="wechat-voice-save-list" style="max-height: 300px; overflow-y: auto;">
+          <!-- 语音列表将动态生成 -->
+        </div>
+        <div class="wechat-modal-actions" style="margin-top: 16px; display: flex; gap: 10px;">
+          <button class="wechat-btn" id="wechat-voice-save-skip" style="flex: 1;">不保存</button>
+          <button class="wechat-btn wechat-btn-primary" id="wechat-voice-save-confirm" style="flex: 1;">保存选中</button>
+        </div>
+      </div>
+    </div>
   `;
 }
 
@@ -969,12 +1060,14 @@ function generateVoiceCallPageHTML() {
       <!-- 通话中对话框 -->
       <div class="wechat-voice-call-chat hidden" id="wechat-voice-call-chat">
         <div class="wechat-voice-call-messages" id="wechat-voice-call-messages"></div>
-        <div class="wechat-voice-call-input-area">
-          <input type="text" class="wechat-voice-call-input" id="wechat-voice-call-input" placeholder="输入文字...">
-          <button class="wechat-voice-call-send" id="wechat-voice-call-send">
-            <svg viewBox="0 0 24 24" fill="currentColor"><path d="M2.01 21L23 12 2.01 3 2 10l15 2-15 2z"/></svg>
-          </button>
-        </div>
+      </div>
+
+      <!-- 通话输入框（独立于对话框，放在按钮上方） -->
+      <div class="wechat-voice-call-input-area hidden" id="wechat-voice-call-input-area">
+        <input type="text" class="wechat-voice-call-input" id="wechat-voice-call-input" placeholder="输入文字...">
+        <button class="wechat-voice-call-send" id="wechat-voice-call-send">
+          <svg viewBox="0 0 24 24" fill="currentColor"><path d="M2.01 21L23 12 2.01 3 2 10l15 2-15 2z"/></svg>
+        </button>
       </div>
 
       <!-- 来电接听按钮（AI发起时显示） -->
@@ -1050,12 +1143,14 @@ function generateVideoCallPageHTML() {
       <!-- 通话中对话框 -->
       <div class="wechat-video-call-chat hidden" id="wechat-video-call-chat">
         <div class="wechat-video-call-messages" id="wechat-video-call-messages"></div>
-        <div class="wechat-video-call-input-area">
-          <input type="text" class="wechat-video-call-input" id="wechat-video-call-input" placeholder="输入文字...">
-          <button class="wechat-video-call-send" id="wechat-video-call-send">
-            <svg viewBox="0 0 24 24" fill="currentColor"><path d="M2.01 21L23 12 2.01 3 2 10l15 2-15 2z"/></svg>
-          </button>
-        </div>
+      </div>
+
+      <!-- 通话输入框（独立于对话框，放在按钮上方） -->
+      <div class="wechat-video-call-input-area hidden" id="wechat-video-call-input-area">
+        <input type="text" class="wechat-video-call-input" id="wechat-video-call-input" placeholder="输入文字...">
+        <button class="wechat-video-call-send" id="wechat-video-call-send">
+          <svg viewBox="0 0 24 24" fill="currentColor"><path d="M2.01 21L23 12 2.01 3 2 10l15 2-15 2z"/></svg>
+        </button>
       </div>
 
       <!-- 底部操作栏 -->
@@ -1113,6 +1208,82 @@ function generateVideoCallPageHTML() {
   `;
 }
 
+// 实时语音通话页面 HTML
+function generateRealVoiceCallPageHTML() {
+  return `
+    <!-- 实时语音通话页面 -->
+    <div id="wechat-real-voice-call-page" class="wechat-real-voice-call-page hidden">
+      <div class="wechat-real-voice-call-header">
+        <button class="wechat-real-voice-call-minimize" id="wechat-real-voice-call-minimize">
+          <svg viewBox="0 0 24 24" width="24" height="24"><rect x="3" y="3" width="18" height="18" rx="2" stroke="currentColor" stroke-width="1.5" fill="none"/><path d="M9 3v18M3 9h6" stroke="currentColor" stroke-width="1.5"/></svg>
+        </button>
+        <span class="wechat-real-voice-call-time hidden" id="wechat-real-voice-call-time">00:00</span>
+        <span style="width: 24px;"></span>
+      </div>
+
+      <div class="wechat-real-voice-call-content">
+        <div class="wechat-real-voice-call-avatar" id="wechat-real-voice-call-avatar"></div>
+        <div class="wechat-real-voice-call-name" id="wechat-real-voice-call-name"></div>
+        <div class="wechat-real-voice-call-status" id="wechat-real-voice-call-status">等待对方接受邀请</div>
+      </div>
+
+      <!-- 通话中消息显示区域 -->
+      <div class="wechat-real-voice-call-chat hidden" id="wechat-real-voice-call-chat">
+        <div class="wechat-real-voice-call-messages" id="wechat-real-voice-call-messages"></div>
+      </div>
+
+      <!-- 说话按钮区域 -->
+      <div class="wechat-real-voice-call-talk-area hidden" id="wechat-real-voice-call-talk-area">
+        <div class="wechat-real-voice-call-talk-btn" id="wechat-real-voice-call-talk-btn">点击 说话</div>
+        <div class="wechat-real-voice-call-talk-hint">点击开始，再点击发送</div>
+        <!-- 文字输入区域（不支持录音时使用） -->
+        <div class="wechat-real-voice-call-text-input-area" id="wechat-real-voice-call-text-input-area">
+          <input type="text" class="wechat-real-voice-call-text-input" id="wechat-real-voice-call-text-input" placeholder="打字说话...">
+          <button class="wechat-real-voice-call-text-send" id="wechat-real-voice-call-text-send">发送</button>
+        </div>
+      </div>
+
+      <!-- 来电接听按钮（AI发起时显示） -->
+      <div class="wechat-real-voice-call-incoming-actions hidden" id="wechat-real-voice-call-incoming-actions">
+        <div class="wechat-real-voice-call-action" id="wechat-real-voice-call-reject">
+          <div class="wechat-real-voice-call-action-btn reject">
+            <svg viewBox="0 0 24 24" width="28" height="28"><path d="M12 9c-1.6 0-3.15.25-4.6.72v3.1c0 .39-.23.74-.56.9-.98.49-1.87 1.12-2.66 1.85-.18.18-.43.28-.7.28-.28 0-.53-.11-.71-.29L.29 13.08a.956.956 0 01-.29-.7c0-.28.11-.53.29-.71C3.34 8.78 7.46 7 12 7s8.66 1.78 11.71 4.67c.18.18.29.43.29.71 0 .28-.11.53-.29.71l-2.48 2.48c-.18.18-.43.29-.71.29-.27 0-.52-.11-.7-.28-.79-.74-1.69-1.36-2.67-1.85-.33-.16-.56-.5-.56-.9v-3.1C15.15 9.25 13.6 9 12 9z" fill="currentColor"/></svg>
+          </div>
+          <span class="wechat-real-voice-call-action-label">拒绝</span>
+        </div>
+        <div class="wechat-real-voice-call-action" id="wechat-real-voice-call-accept">
+          <div class="wechat-real-voice-call-action-btn accept">
+            <svg viewBox="0 0 24 24" width="28" height="28"><path d="M12 1a4 4 0 00-4 4v7a4 4 0 008 0V5a4 4 0 00-4-4z" stroke="currentColor" stroke-width="1.5" fill="none"/><path d="M19 10v2a7 7 0 01-14 0v-2M12 19v4M8 23h8" stroke="currentColor" stroke-width="1.5" stroke-linecap="round" fill="none"/></svg>
+          </div>
+          <span class="wechat-real-voice-call-action-label">接听</span>
+        </div>
+      </div>
+
+      <!-- 通话中操作按钮（接通后显示） -->
+      <div class="wechat-real-voice-call-actions hidden" id="wechat-real-voice-call-actions">
+        <div class="wechat-real-voice-call-action" id="wechat-real-voice-call-mute">
+          <div class="wechat-real-voice-call-action-btn">
+            <svg viewBox="0 0 24 24" width="28" height="28"><path d="M12 1a4 4 0 00-4 4v7a4 4 0 008 0V5a4 4 0 00-4-4z" stroke="currentColor" stroke-width="1.5" fill="none"/><path d="M19 10v2a7 7 0 01-14 0v-2M12 19v4M8 23h8" stroke="currentColor" stroke-width="1.5" stroke-linecap="round" fill="none"/></svg>
+          </div>
+          <span class="wechat-real-voice-call-action-label">静音</span>
+        </div>
+        <div class="wechat-real-voice-call-action" id="wechat-real-voice-call-hangup">
+          <div class="wechat-real-voice-call-action-btn hangup">
+            <svg viewBox="0 0 24 24" width="28" height="28"><path d="M12 9c-1.6 0-3.15.25-4.6.72v3.1c0 .39-.23.74-.56.9-.98.49-1.87 1.12-2.66 1.85-.18.18-.43.28-.7.28-.28 0-.53-.11-.71-.29L.29 13.08a.956.956 0 01-.29-.7c0-.28.11-.53.29-.71C3.34 8.78 7.46 7 12 7s8.66 1.78 11.71 4.67c.18.18.29.43.29.71 0 .28-.11.53-.29.71l-2.48 2.48c-.18.18-.43.29-.71.29-.27 0-.52-.11-.7-.28-.79-.74-1.69-1.36-2.67-1.85-.33-.16-.56-.5-.56-.9v-3.1C15.15 9.25 13.6 9 12 9z" fill="currentColor"/></svg>
+          </div>
+          <span class="wechat-real-voice-call-action-label">挂断</span>
+        </div>
+        <div class="wechat-real-voice-call-action" id="wechat-real-voice-call-speaker">
+          <div class="wechat-real-voice-call-action-btn">
+            <svg viewBox="0 0 24 24" width="28" height="28"><path d="M11 5L6 9H2v6h4l5 4V5z" stroke="currentColor" stroke-width="1.5" fill="none"/><path d="M15.54 8.46a5 5 0 010 7.07M19.07 4.93a10 10 0 010 14.14" stroke="currentColor" stroke-width="1.5" stroke-linecap="round" fill="none"/></svg>
+          </div>
+          <span class="wechat-real-voice-call-action-label">扬声器</span>
+        </div>
+      </div>
+    </div>
+  `;
+}
+
 // 朋友圈页面 HTML
 function generateMomentsPageHTML() {
   return `
@@ -1654,11 +1825,12 @@ function generateHistoryPageHTML() {
         <div style="width: 24px;"></div>
       </div>
 
-      <!-- 四个标签按钮 -->
+      <!-- 五个标签按钮 -->
       <div class="wechat-history-tabs">
         <button class="wechat-history-tab active" data-tab="listen">一起听</button>
         <button class="wechat-history-tab" data-tab="voice">语音通话</button>
         <button class="wechat-history-tab" data-tab="video">视频通话</button>
+        <button class="wechat-history-tab wechat-history-tab-green" data-tab="playback">语音回放</button>
         <button class="wechat-history-tab wechat-history-tab-pink" data-tab="toy">心动瞬间</button>
       </div>
 

style.css

@@ -4888,7 +4888,11 @@
   background: rgba(50, 50, 50, 0.8);
   border-radius: 20px;
   padding: 4px 4px 4px 16px;
-  margin-top: auto;
+  margin: 0 16px 15px 16px;
+}
+
+.wechat-voice-call-input-area.hidden {
+  display: none;
 }
 
 .wechat-voice-call-input {
@@ -5300,6 +5304,11 @@
   display: flex;
   gap: 8px;
   padding: 8px 0;
+  margin: 0 16px 10px 16px;
+}
+
+.wechat-video-call-input-area.hidden {
+  display: none;
 }
 
 .wechat-video-call-input {
@@ -5467,6 +5476,573 @@
   transform: scale(1.05);
 }
 
+/* ===== 实时语音通话页面 ===== */
+.wechat-real-voice-call-page {
+  position: absolute;
+  top: 0;
+  left: 0;
+  right: 0;
+  bottom: 0;
+  background: linear-gradient(135deg, #1a1a2e 0%, #16213e 50%, #0f3460 100%);
+  z-index: 1100;
+  display: flex;
+  flex-direction: column;
+  overflow: hidden;
+}
+
+.wechat-real-voice-call-page.hidden {
+  display: none;
+}
+
+.wechat-real-voice-call-header {
+  display: flex;
+  align-items: center;
+  justify-content: space-between;
+  padding: 16px;
+  background: transparent;
+  position: relative;
+  z-index: 10;
+}
+
+.wechat-real-voice-call-minimize {
+  width: 36px;
+  height: 36px;
+  border-radius: 50%;
+  background: rgba(255,255,255,0.15);
+  border: none;
+  color: #fff;
+  display: flex;
+  align-items: center;
+  justify-content: center;
+  cursor: pointer;
+  transition: all 0.2s;
+}
+
+.wechat-real-voice-call-minimize:hover {
+  background: rgba(255,255,255,0.25);
+}
+
+.wechat-real-voice-call-time {
+  font-size: 16px;
+  color: #fff;
+  font-weight: 500;
+}
+
+.wechat-real-voice-call-time.hidden {
+  visibility: hidden;
+}
+
+.wechat-real-voice-call-content {
+  flex: 1;
+  display: flex;
+  flex-direction: column;
+  align-items: center;
+  padding: 20px;
+  overflow: hidden;
+}
+
+.wechat-real-voice-call-avatar {
+  width: 100px;
+  height: 100px;
+  border-radius: 50%;
+  background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
+  display: flex;
+  align-items: center;
+  justify-content: center;
+  font-size: 40px;
+  color: #fff;
+  margin-bottom: 16px;
+  box-shadow: 0 8px 32px rgba(0,0,0,0.3);
+  overflow: hidden;
+}
+
+.wechat-real-voice-call-avatar img {
+  width: 100%;
+  height: 100%;
+  object-fit: cover;
+}
+
+.wechat-real-voice-call-name {
+  font-size: 22px;
+  color: #fff;
+  font-weight: 600;
+  margin-bottom: 8px;
+  text-shadow: 0 2px 8px rgba(0,0,0,0.3);
+}
+
+.wechat-real-voice-call-status {
+  font-size: 14px;
+  color: rgba(255,255,255,0.7);
+  margin-bottom: 20px;
+}
+
+.wechat-real-voice-call-status.connecting {
+  animation: pulse 1.5s ease-in-out infinite;
+}
+
+@keyframes pulse {
+  0%, 100% { opacity: 0.7; }
+  50% { opacity: 1; }
+}
+
+/* 实时语音对话区域 */
+.wechat-real-voice-call-chat {
+  flex: 1;
+  display: flex;
+  flex-direction: column;
+  width: 100%;
+  max-width: 320px;
+  margin: 0 auto;
+  background: rgba(255,255,255,0.05);
+  border-radius: 16px;
+  overflow: hidden;
+  backdrop-filter: blur(10px);
+}
+
+.wechat-real-voice-call-chat.hidden {
+  display: none;
+}
+
+.wechat-real-voice-call-messages {
+  flex: 1;
+  overflow-y: auto;
+  padding: 16px;
+  display: flex;
+  flex-direction: column;
+  gap: 12px;
+  min-height: 120px;
+  max-height: 200px;
+}
+
+.wechat-real-voice-call-messages::-webkit-scrollbar {
+  width: 4px;
+}
+
+.wechat-real-voice-call-messages::-webkit-scrollbar-thumb {
+  background: rgba(255,255,255,0.2);
+  border-radius: 2px;
+}
+
+.wechat-real-voice-call-msg {
+  max-width: 85%;
+  padding: 10px 14px;
+  border-radius: 12px;
+  font-size: 14px;
+  line-height: 1.5;
+  word-break: break-word;
+}
+
+.wechat-real-voice-call-msg.ai {
+  align-self: flex-start;
+  background: rgba(255,255,255,0.15);
+  color: #fff;
+  border-bottom-left-radius: 4px;
+}
+
+.wechat-real-voice-call-msg.user {
+  align-self: flex-end;
+  background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
+  color: #fff;
+  border-bottom-right-radius: 4px;
+}
+
+.wechat-real-voice-call-msg.fade-in {
+  animation: msgFadeIn 0.3s ease-out;
+}
+
+@keyframes msgFadeIn {
+  from { opacity: 0; transform: translateY(10px); }
+  to { opacity: 1; transform: translateY(0); }
+}
+
+/* 按住说话区域 */
+.wechat-real-voice-call-talk-area {
+  padding: 16px;
+  display: flex;
+  flex-direction: column;
+  align-items: center;
+  gap: 8px;
+}
+
+.wechat-real-voice-call-talk-area.hidden {
+  display: none;
+}
+
+.wechat-real-voice-call-talk-btn {
+  width: 100%;
+  max-width: 280px;
+  padding: 16px 24px;
+  border-radius: 24px;
+  background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
+  border: none;
+  color: #fff;
+  font-size: 16px;
+  font-weight: 500;
+  cursor: pointer;
+  transition: all 0.2s;
+  user-select: none;
+  -webkit-user-select: none;
+  touch-action: none;
+}
+
+.wechat-real-voice-call-talk-btn:hover {
+  transform: scale(1.02);
+  box-shadow: 0 4px 20px rgba(102, 126, 234, 0.4);
+}
+
+.wechat-real-voice-call-talk-btn:active,
+.wechat-real-voice-call-talk-btn.recording {
+  background: linear-gradient(135deg, #ff6b6b 0%, #ee5a5a 100%);
+  transform: scale(0.98);
+}
+
+.wechat-real-voice-call-talk-hint {
+  font-size: 12px;
+  color: rgba(255,255,255,0.5);
+}
+
+/* 文字输入区域（不支持录音时） */
+.wechat-real-voice-call-text-input-area {
+  display: flex;
+  align-items: center;
+  gap: 8px;
+  margin-top: 15px;
+  padding: 0 20px;
+  width: 100%;
+  box-sizing: border-box;
+}
+
+.wechat-real-voice-call-text-input {
+  flex: 1;
+  height: 40px;
+  border: none;
+  border-radius: 20px;
+  padding: 0 16px;
+  font-size: 14px;
+  background: rgba(255,255,255,0.15);
+  color: #fff;
+  outline: none;
+}
+
+.wechat-real-voice-call-text-input::placeholder {
+  color: rgba(255,255,255,0.5);
+}
+
+.wechat-real-voice-call-text-send {
+  height: 40px;
+  padding: 0 20px;
+  border: none;
+  border-radius: 20px;
+  background: #07c160;
+  color: #fff;
+  font-size: 14px;
+  font-weight: 500;
+  cursor: pointer;
+}
+
+.wechat-real-voice-call-text-send:active {
+  background: #06ad56;
+}
+
+/* 操作按钮区域 */
+.wechat-real-voice-call-actions {
+  display: flex;
+  justify-content: center;
+  gap: 48px;
+  padding: 24px;
+  background: transparent;
+}
+
+.wechat-real-voice-call-action {
+  display: flex;
+  flex-direction: column;
+  align-items: center;
+  gap: 8px;
+}
+
+.wechat-real-voice-call-action-btn {
+  width: 56px;
+  height: 56px;
+  border-radius: 50%;
+  background: rgba(255,255,255,0.15);
+  border: none;
+  color: #fff;
+  display: flex;
+  align-items: center;
+  justify-content: center;
+  cursor: pointer;
+  transition: all 0.2s;
+}
+
+.wechat-real-voice-call-action-btn:hover {
+  background: rgba(255,255,255,0.25);
+  transform: scale(1.05);
+}
+
+.wechat-real-voice-call-action-btn.hangup {
+  background: #ff4444;
+}
+
+.wechat-real-voice-call-action-btn.hangup:hover {
+  background: #ff6666;
+}
+
+.wechat-real-voice-call-action-btn.muted {
+  background: rgba(255,68,68,0.3);
+}
+
+.wechat-real-voice-call-action-btn.muted svg {
+  opacity: 0.7;
+}
+
+.wechat-real-voice-call-action-label {
+  font-size: 12px;
+  color: rgba(255,255,255,0.7);
+}
+
+/* 来电操作按钮 */
+.wechat-real-voice-call-incoming-actions {
+  display: flex;
+  justify-content: center;
+  gap: 60px;
+  padding: 24px;
+}
+
+.wechat-real-voice-call-incoming-actions.hidden {
+  display: none;
+}
+
+.wechat-real-voice-call-action-btn.reject {
+  background: #ff4444;
+}
+
+.wechat-real-voice-call-action-btn.reject:hover {
+  background: #ff6666;
+}
+
+.wechat-real-voice-call-action-btn.accept {
+  background: #07c160;
+}
+
+.wechat-real-voice-call-action-btn.accept:hover {
+  background: #1ed76a;
+}
+
+/* 实时语音通话记录样式 */
+.wechat-real-voice-record .wechat-call-record-icon {
+  transform: none;
+}
+
+/* ===== 语音 API 设置面板 ===== */
+.wechat-voice-api-panel {
+  position: absolute;
+  left: 0;
+  right: 0;
+  bottom: 0;
+  background: var(--wechat-bg);
+  border-radius: 16px 16px 0 0;
+  box-shadow: 0 -4px 20px rgba(0,0,0,0.15);
+  z-index: 200;
+  max-height: 80%;
+  overflow-y: auto;
+  animation: slideUp 0.3s ease-out;
+}
+
+.wechat-voice-api-panel.hidden {
+  display: none;
+}
+
+@keyframes slideUp {
+  from { transform: translateY(100%); }
+  to { transform: translateY(0); }
+}
+
+.wechat-voice-api-header {
+  display: flex;
+  align-items: center;
+  justify-content: space-between;
+  padding: 16px;
+  border-bottom: 1px solid var(--wechat-border);
+  position: sticky;
+  top: 0;
+  background: var(--wechat-bg);
+  z-index: 10;
+}
+
+.wechat-voice-api-title {
+  font-size: 16px;
+  font-weight: 600;
+  color: var(--wechat-text-primary);
+}
+
+.wechat-voice-api-close {
+  width: 28px;
+  height: 28px;
+  border-radius: 50%;
+  background: var(--wechat-border);
+  border: none;
+  color: var(--wechat-text-secondary);
+  display: flex;
+  align-items: center;
+  justify-content: center;
+  cursor: pointer;
+  font-size: 16px;
+}
+
+.wechat-voice-api-close:hover {
+  background: var(--wechat-hover);
+}
+
+.wechat-voice-api-content {
+  padding: 16px;
+  display: flex;
+  flex-direction: column;
+  gap: 20px;
+}
+
+.wechat-voice-api-section {
+  background: var(--wechat-card-bg, #fff);
+  border-radius: 12px;
+  padding: 16px;
+  box-shadow: 0 1px 3px rgba(0,0,0,0.05);
+}
+
+.wechat-dark .wechat-voice-api-section {
+  background: var(--wechat-card-bg, #1e1e1e);
+}
+
+.wechat-voice-api-section-title {
+  font-size: 14px;
+  font-weight: 600;
+  color: var(--wechat-text-primary);
+  margin-bottom: 12px;
+  display: flex;
+  align-items: center;
+  gap: 8px;
+}
+
+.wechat-voice-api-section-title svg {
+  width: 18px;
+  height: 18px;
+  opacity: 0.7;
+}
+
+.wechat-voice-api-row {
+  display: flex;
+  flex-direction: column;
+  gap: 6px;
+  margin-bottom: 12px;
+}
+
+.wechat-voice-api-row:last-child {
+  margin-bottom: 0;
+}
+
+.wechat-voice-api-label {
+  font-size: 12px;
+  color: var(--wechat-text-secondary);
+}
+
+.wechat-voice-api-input-group {
+  display: flex;
+  gap: 8px;
+}
+
+.wechat-voice-api-input {
+  flex: 1;
+  padding: 10px 12px;
+  border: 1px solid var(--wechat-border);
+  border-radius: 8px;
+  font-size: 14px;
+  background: var(--wechat-bg);
+  color: var(--wechat-text-primary);
+}
+
+.wechat-voice-api-input:focus {
+  outline: none;
+  border-color: var(--wechat-primary);
+}
+
+.wechat-voice-api-input::placeholder {
+  color: var(--wechat-text-secondary);
+  opacity: 0.6;
+}
+
+.wechat-voice-api-eye-btn {
+  width: 40px;
+  height: 40px;
+  border: 1px solid var(--wechat-border);
+  border-radius: 8px;
+  background: var(--wechat-bg);
+  color: var(--wechat-text-secondary);
+  display: flex;
+  align-items: center;
+  justify-content: center;
+  cursor: pointer;
+}
+
+.wechat-voice-api-eye-btn:hover {
+  background: var(--wechat-hover);
+}
+
+.wechat-voice-api-test-btn {
+  padding: 10px 16px;
+  border: 1px solid var(--wechat-primary);
+  border-radius: 8px;
+  background: transparent;
+  color: var(--wechat-primary);
+  font-size: 14px;
+  cursor: pointer;
+  transition: all 0.2s;
+}
+
+.wechat-voice-api-test-btn:hover {
+  background: var(--wechat-primary);
+  color: #fff;
+}
+
+.wechat-voice-api-test-btn:disabled {
+  opacity: 0.5;
+  cursor: not-allowed;
+}
+
+.wechat-voice-api-row-inline {
+  display: flex;
+  gap: 12px;
+}
+
+.wechat-voice-api-row-inline .wechat-voice-api-row {
+  flex: 1;
+  margin-bottom: 0;
+}
+
+.wechat-voice-api-actions {
+  display: flex;
+  gap: 12px;
+  padding: 16px;
+  border-top: 1px solid var(--wechat-border);
+  background: var(--wechat-bg);
+  position: sticky;
+  bottom: 0;
+}
+
+.wechat-voice-api-save-btn {
+  flex: 1;
+  padding: 12px;
+  border: none;
+  border-radius: 8px;
+  background: var(--wechat-primary);
+  color: #fff;
+  font-size: 15px;
+  font-weight: 500;
+  cursor: pointer;
+  transition: all 0.2s;
+}
+
+.wechat-voice-api-save-btn:hover {
+  opacity: 0.9;
+}
+
 /* ===== 音乐搜索面板 ===== */
 .wechat-music-panel {
   position: absolute;
@@ -11951,6 +12527,114 @@
   opacity: 1;
 }
 
+/* ========== 语音回放标签样式 ========== */
+.wechat-history-tab-green {
+  background: linear-gradient(135deg, #07c160, #06ad56) !important;
+}
+
+.wechat-history-tab-green.active {
+  opacity: 1;
+}
+
+/* 语音回放列表 */
+.wechat-voice-playback-list {
+  display: flex;
+  flex-direction: column;
+  gap: 12px;
+  padding: 12px;
+}
+
+.wechat-voice-playback-card {
+  background: linear-gradient(135deg, #f0fff4, #e6ffed);
+  border-radius: 12px;
+  padding: 12px;
+  box-shadow: 0 2px 8px rgba(7, 193, 96, 0.15);
+}
+
+.wechat-dark .wechat-voice-playback-card {
+  background: linear-gradient(135deg, #1a3025, #0d2818);
+}
+
+.wechat-voice-playback-card-header {
+  display: flex;
+  justify-content: space-between;
+  align-items: center;
+  margin-bottom: 8px;
+}
+
+.wechat-voice-playback-time {
+  font-size: 12px;
+  color: var(--wechat-text-secondary);
+}
+
+.wechat-voice-playback-actions {
+  display: flex;
+  align-items: center;
+  gap: 8px;
+}
+
+.wechat-voice-playback-duration {
+  font-size: 12px;
+  color: #07c160;
+  font-weight: 500;
+}
+
+.wechat-voice-playback-delete {
+  width: 20px;
+  height: 20px;
+  border: none;
+  background: rgba(255, 77, 79, 0.1);
+  color: #ff4d4f;
+  border-radius: 50%;
+  cursor: pointer;
+  font-size: 14px;
+  display: flex;
+  align-items: center;
+  justify-content: center;
+}
+
+.wechat-voice-playback-delete:hover {
+  background: rgba(255, 77, 79, 0.2);
+}
+
+.wechat-voice-playback-content {
+  display: flex;
+  align-items: center;
+  gap: 10px;
+}
+
+.wechat-voice-playback-text {
+  flex: 1;
+  font-size: 13px;
+  color: var(--wechat-text-primary);
+  line-height: 1.5;
+  word-break: break-word;
+}
+
+.wechat-voice-playback-btn {
+  width: 36px;
+  height: 36px;
+  border-radius: 50%;
+  background: #07c160;
+  border: none;
+  color: #fff;
+  display: flex;
+  align-items: center;
+  justify-content: center;
+  cursor: pointer;
+  flex-shrink: 0;
+  transition: background 0.2s;
+}
+
+.wechat-voice-playback-btn:hover {
+  background: #06ad56;
+}
+
+.wechat-voice-playback-btn:disabled {
+  background: #aaa;
+  cursor: not-allowed;
+}
+
 /* 心动瞬间卡片样式 */
 .wechat-toy-history-card {
   background: linear-gradient(135deg, #fff5f8, #ffe4ec);
@@ -12207,3 +12891,151 @@
   transition: transform 0.3s cubic-bezier(0.4, 0, 0.2, 1),
               box-shadow 0.3s ease;
 }
+
+/* 语音回放保存弹窗 */
+.wechat-voice-save-list {
+  display: flex;
+  flex-direction: column;
+  gap: 8px;
+}
+
+.wechat-voice-save-item {
+  display: flex;
+  align-items: center;
+  gap: 10px;
+  padding: 10px 12px;
+  background: var(--wechat-bg-light);
+  border-radius: 8px;
+  cursor: pointer;
+  transition: background 0.2s;
+}
+
+.wechat-voice-save-item:hover {
+  background: var(--wechat-bg-hover);
+}
+
+.wechat-voice-save-item.selected {
+  background: rgba(7, 193, 96, 0.15);
+  border: 1px solid #07c160;
+}
+
+.wechat-voice-save-checkbox {
+  position: relative;
+  width: 18px;
+  height: 18px;
+  flex-shrink: 0;
+}
+
+.wechat-voice-save-checkbox input[type="checkbox"] {
+  position: absolute;
+  opacity: 0;
+  width: 100%;
+  height: 100%;
+  cursor: pointer;
+  z-index: 1;
+}
+
+.wechat-voice-save-checkbox label {
+  position: absolute;
+  top: 0;
+  left: 0;
+  width: 18px;
+  height: 18px;
+  border: 2px solid var(--wechat-border);
+  border-radius: 4px;
+  background: transparent;
+  transition: all 0.2s;
+}
+
+.wechat-voice-save-checkbox input[type="checkbox"]:checked + label {
+  background: #07c160;
+  border-color: #07c160;
+}
+
+.wechat-voice-save-checkbox label::after {
+  content: '';
+  position: absolute;
+  top: 2px;
+  left: 5px;
+  width: 4px;
+  height: 8px;
+  border: solid #fff;
+  border-width: 0 2px 2px 0;
+  transform: rotate(45deg);
+  opacity: 0;
+  transition: opacity 0.2s;
+}
+
+.wechat-voice-save-checkbox input[type="checkbox"]:checked + label::after {
+  opacity: 1;
+}
+
+.wechat-voice-save-info {
+  flex: 1;
+  min-width: 0;
+}
+
+.wechat-voice-save-text {
+  flex: 1;
+  font-size: 13px;
+  color: var(--wechat-text-primary);
+  overflow: hidden;
+  text-overflow: ellipsis;
+  white-space: nowrap;
+}
+
+.wechat-voice-save-play {
+  width: 28px;
+  height: 28px;
+  border-radius: 50%;
+  background: #07c160;
+  border: none;
+  color: #fff;
+  display: flex;
+  align-items: center;
+  justify-content: center;
+  cursor: pointer;
+  flex-shrink: 0;
+}
+
+.wechat-voice-save-play:hover {
+  background: #06ad56;
+}
+
+.wechat-voice-save-play svg {
+  width: 14px;
+  height: 14px;
+}
+
+.wechat-voice-save-duration {
+  font-size: 12px;
+  color: var(--wechat-text-secondary);
+  flex-shrink: 0;
+}
+
+/* 语音回放气泡 */
+.wechat-voice-playback {
+  display: flex;
+  align-items: center;
+  gap: 8px;
+  margin-top: 6px;
+  padding: 6px 10px;
+  background: rgba(7, 193, 96, 0.1);
+  border-radius: 6px;
+  cursor: pointer;
+}
+
+.wechat-voice-playback:hover {
+  background: rgba(7, 193, 96, 0.2);
+}
+
+.wechat-voice-playback-icon {
+  width: 20px;
+  height: 20px;
+  color: #07c160;
+}
+
+.wechat-voice-playback-text {
+  font-size: 12px;
+  color: #07c160;
+}

summary.js

@@ -311,23 +311,23 @@ export function generateSummaryPrompt(allChats, cupNumber) {
     prompt = settings.customSummaryTemplate.trim() + '\n\n';
   } else {
     // 使用默认模板（纯对话记录模式）
-    prompt = `你的任务是将这段【线上聊天记录】原样整理成JSON格式。
+    prompt = `【重要】你必须且只能输出一个JSON对象，禁止输出任何其他内容。
 
-【核心原则】
-- 原样保留：完整复制每一条对话，不做任何修改、润色或总结
-- 格式统一：按"发言者: 内容"格式逐行记录
-- 仅提取关键词：从对话中提取3-5个核心关键词用于检索触发
+你的任务是将聊天记录整理成JSON格式。
 
-【输出格式要求】
-- 只输出一个JSON对象
-- 不要使用markdown代码块
-- 直接以 { 开头，以 } 结尾
-- keys: 3-5个能代表本次聊天核心内容的关键词（人名、地点、事件等）
-- content: 以"以下是线上聊天内容："开头，然后原样复制对话记录，每条一行，格式为"发言者: 内容"
-- comment: "${getCupName(cupNumber)}"
+【输出规则 - 必须严格遵守】
+1. 直接以 { 开头，以 } 结尾
+2. 禁止使用markdown代码块（禁止\`\`\`）
+3. 禁止输出任何解释、思考、前言
+4. 禁止在JSON前后添加任何文字
 
-【JSON示例】
-{"keys":["公园","约会","周末"],"content":"以下是线上聊天内容：\\n{{user}}: 今天去哪玩？\\n{{char}}: 去公园吧\\n{{user}}: 好呀\\n{{char}}: 那我们下午2点见","comment":"${getCupName(cupNumber)}"}
+【JSON字段说明】
+- "keys": 数组，3-5个关键词（人名、地点、事件等）
+- "content": 字符串，以"以下是线上聊天内容："开头，然后逐行记录对话，格式为"发言者: 内容"，用\\n分隔
+- "comment": "${getCupName(cupNumber)}"
+
+【正确输出示例】
+{"keys":["公园","约会","周末"],"content":"以下是线上聊天内容：\\n{{user}}: 今天去哪玩？\\n{{char}}: 去公园吧","comment":"${getCupName(cupNumber)}"}
 
 `;
   }
@@ -469,13 +469,13 @@ export function generateSummaryPrompt(allChats, cupNumber) {
     });
   });
 
-  prompt += `\n请将以上聊天记录原样整理成${getCupName(cupNumber)}的JSON：`;
+  prompt += `\n【立即输出JSON】请将以上聊天记录整理成${getCupName(cupNumber)}的JSON对象（直接以{开头）：`;
 
   return prompt;
 }
 
 // 调用总结API
-export async function callSummaryAPI(prompt) {
+export async function callSummaryAPI(prompt, cupNumber = 1) {
   const settings = getSettings();
   const apiUrl = settings.summaryApiUrl;
   const apiKey = settings.summaryApiKey;
@@ -513,14 +513,16 @@ export async function callSummaryAPI(prompt) {
   const content = data.choices?.[0]?.message?.content || '';
 
   // 解析JSON
-  const parsed = parseJSONResponse(content);
+  const parsed = parseJSONResponse(content, cupNumber);
   if (parsed) return parsed;
 
   throw new Error('AI返回内容为空或无法解析');
 }
 
 // 解析JSON响应
-function parseJSONResponse(content) {
+function parseJSONResponse(content, cupNumber = 1) {
+  if (!content || !content.trim()) return null;
+
   // 方法1: 直接解析
   try {
     const result = JSON.parse(content);
@@ -545,16 +547,26 @@ function parseJSONResponse(content) {
     }
   } catch (e) {}
 
-  // 降级方案
-  if (content && content.trim().length > 20) {
-    const words = content.match(/[\u4e00-\u9fa5]{2,}/g) || ['聊天', '记录'];
-    return {
-      keys: [...new Set(words)].slice(0, 5),
-      content: content.substring(0, 30000).replace(/```[\s\S]*?```/g, '').trim(),
-      comment: '感情记录'
-    };
-  }
+  // 方法4: 尝试修复常见的JSON格式问题
+  try {
+    // 替换中文冒号和引号
+    let fixed = content
+      .replace(/：/g, ':')
+      .replace(/"/g, '"')
+      .replace(/"/g, '"')
+      .replace(/'/g, "'")
+      .replace(/'/g, "'");
 
+    const firstBrace = fixed.indexOf('{');
+    const lastBrace = fixed.lastIndexOf('}');
+    if (firstBrace !== -1 && lastBrace > firstBrace) {
+      const result = JSON.parse(fixed.substring(firstBrace, lastBrace + 1));
+      if (result.keys && result.content) return result;
+    }
+  } catch (e) {}
+
+  // 不再使用降级方案，返回null让调用者处理错误
+  console.error('[可乐] JSON解析失败，原始内容前500字符:', content.substring(0, 500));
   return null;
 }
 
@@ -732,7 +744,7 @@ export async function executeSummary() {
         // 为单个聊天生成总结
         updateProgress('🤖 分析 ' + chat.contactName + ' 的' + getCupName(cupNumber) + '...');
         const prompt = generateSummaryPrompt([chat], cupNumber);
-        const entry = await callSummaryAPI(prompt);
+        const entry = await callSummaryAPI(prompt, cupNumber);
 
         // 保存到收藏
         saveEntryToFavorites(entry, cupNumber, lorebookName);

video-call.js

@@ -101,6 +101,7 @@ function showIncomingCallPage() {
   // 隐藏主界面元素，显示来电界面
   document.getElementById('wechat-video-call-center')?.classList.add('hidden');
   document.getElementById('wechat-video-call-chat')?.classList.add('hidden');
+  document.getElementById('wechat-video-call-input-area')?.classList.add('hidden');
   document.getElementById('wechat-video-call-actions')?.classList.add('hidden');
   incomingEl.classList.remove('hidden');
 
@@ -165,8 +166,9 @@ function showCallPage() {
     timeEl.classList.add('hidden'); // 拨打中不显示计时
   }
 
-  // 隐藏对话框
+  // 隐藏对话框和输入框
   document.getElementById('wechat-video-call-chat')?.classList.add('hidden');
+  document.getElementById('wechat-video-call-input-area')?.classList.add('hidden');
   document.getElementById('wechat-video-call-messages')?.innerHTML &&
     (document.getElementById('wechat-video-call-messages').innerHTML = '');
 
@@ -266,8 +268,9 @@ function onVideoCallConnected() {
   document.getElementById('wechat-video-call-incoming')?.classList.add('hidden');
   document.getElementById('wechat-video-call-actions')?.classList.remove('hidden');
 
-  // 显示对话框
+  // 显示对话框和输入框
   document.getElementById('wechat-video-call-chat')?.classList.remove('hidden');
+  document.getElementById('wechat-video-call-input-area')?.classList.remove('hidden');
 
   // 接通后才显示计时
   const timeEl = document.getElementById('wechat-video-call-time');

voice-api.js

@@ -0,0 +1,731 @@
+/**
+ * 语音 API 封装
+ * TTS (文字转语音) 和 STT (语音转文字)
+ */
+
+import { getSettings } from './config.js';
+
+/**
+ * 获取语音 API 配置
+ * @param {Object} contact - 角色对象（可选，用于获取角色独立配置）
+ * @returns {Object} 配置对象
+ */
+export function getVoiceApiConfig(contact = null) {
+  const settings = getSettings();
+
+  // 基础配置
+  const config = {
+    stt: {
+      url: settings.sttApiUrl || '',
+      key: settings.sttApiKey || '',
+      model: settings.sttModel || ''
+    },
+    tts: {
+      url: settings.ttsApiUrl || '',
+      key: settings.ttsApiKey || '',
+      model: settings.ttsModel || '',
+      voice: settings.ttsVoice || '',
+      speed: settings.ttsSpeed || 1,
+      emotion: settings.ttsEmotion || '默认',
+      proxyUrl: settings.ttsProxyUrl || ''
+    }
+  };
+
+  // 角色独立 TTS 配置
+  if (contact?.useCustomVoice && contact.customTtsVoice) {
+    config.tts.voice = contact.customTtsVoice;
+  }
+
+  return config;
+}
+
+/**
+ * 根据 Blob 类型获取文件名
+ */
+function getAudioFileName(blob) {
+  const type = blob.type || 'audio/webm';
+  if (type.includes('webm')) return 'audio.webm';
+  if (type.includes('ogg')) return 'audio.ogg';
+  if (type.includes('mp4')) return 'audio.mp4';
+  if (type.includes('mpeg') || type.includes('mp3')) return 'audio.mp3';
+  if (type.includes('wav')) return 'audio.wav';
+  if (type.includes('flac')) return 'audio.flac';
+  return 'audio.webm';
+}
+
+/**
+ * 将音频 Blob 转换为 WAV 格式（更好的兼容性）
+ * 导出供其他模块使用
+ */
+export async function convertToWav(audioBlob) {
+  try {
+    const audioContext = new (window.AudioContext || window.webkitAudioContext)();
+    const arrayBuffer = await audioBlob.arrayBuffer();
+    const audioBuffer = await audioContext.decodeAudioData(arrayBuffer);
+
+    // 创建 WAV 文件
+    const numChannels = audioBuffer.numberOfChannels;
+    const sampleRate = audioBuffer.sampleRate;
+    const format = 1; // PCM
+    const bitDepth = 16;
+
+    const bytesPerSample = bitDepth / 8;
+    const blockAlign = numChannels * bytesPerSample;
+
+    const samples = audioBuffer.length;
+    const dataSize = samples * blockAlign;
+    const buffer = new ArrayBuffer(44 + dataSize);
+    const view = new DataView(buffer);
+
+    // WAV 头部
+    const writeString = (offset, str) => {
+      for (let i = 0; i < str.length; i++) {
+        view.setUint8(offset + i, str.charCodeAt(i));
+      }
+    };
+
+    writeString(0, 'RIFF');
+    view.setUint32(4, 36 + dataSize, true);
+    writeString(8, 'WAVE');
+    writeString(12, 'fmt ');
+    view.setUint32(16, 16, true);
+    view.setUint16(20, format, true);
+    view.setUint16(22, numChannels, true);
+    view.setUint32(24, sampleRate, true);
+    view.setUint32(28, sampleRate * blockAlign, true);
+    view.setUint16(32, blockAlign, true);
+    view.setUint16(34, bitDepth, true);
+    writeString(36, 'data');
+    view.setUint32(40, dataSize, true);
+
+    // 写入音频数据
+    const channelData = [];
+    for (let i = 0; i < numChannels; i++) {
+      channelData.push(audioBuffer.getChannelData(i));
+    }
+
+    let offset = 44;
+    for (let i = 0; i < samples; i++) {
+      for (let ch = 0; ch < numChannels; ch++) {
+        const sample = Math.max(-1, Math.min(1, channelData[ch][i]));
+        const intSample = sample < 0 ? sample * 0x8000 : sample * 0x7FFF;
+        view.setInt16(offset, intSample, true);
+        offset += 2;
+      }
+    }
+
+    await audioContext.close();
+    return new Blob([buffer], { type: 'audio/wav' });
+  } catch (err) {
+    console.warn('[可乐] WAV 转换失败，使用原格式:', err);
+    return audioBlob;
+  }
+}
+
+/**
+ * STT: 语音转文字
+ * @param {Blob} audioBlob - 音频数据
+ * @param {Object} options - 选项
+ * @returns {Promise<string>} 识别的文字
+ */
+export async function speechToText(audioBlob, options = {}) {
+  const config = getVoiceApiConfig();
+
+  if (!config.stt.url || !config.stt.key) {
+    throw new Error('请先配置语音识别 (STT) API');
+  }
+
+  // 自动补全 URL 路径
+  let sttUrl = config.stt.url.trim().replace(/\/+$/, '');
+  if (!sttUrl.includes('/audio/transcriptions')) {
+    sttUrl = sttUrl + '/audio/transcriptions';
+  }
+
+  // 如果不是 WAV 格式，尝试转换以提高兼容性
+  let processedBlob = audioBlob;
+  if (!audioBlob.type.includes('wav')) {
+    console.log('[可乐] 转换音频为 WAV 格式...');
+    processedBlob = await convertToWav(audioBlob);
+  }
+
+  // 根据音频类型设置正确的文件名
+  const fileName = getAudioFileName(processedBlob);
+
+  const formData = new FormData();
+  formData.append('file', processedBlob, fileName);
+
+  if (config.stt.model) {
+    formData.append('model', config.stt.model);
+  }
+
+  try {
+    console.log('[可乐] STT 请求:', {
+      url: sttUrl,
+      model: config.stt.model,
+      originalType: audioBlob.type,
+      processedType: processedBlob.type,
+      audioSize: processedBlob.size,
+      fileName: fileName
+    });
+
+    const response = await fetch(sttUrl, {
+      method: 'POST',
+      headers: {
+        'Authorization': `Bearer ${config.stt.key}`
+      },
+      body: formData
+    });
+
+    if (!response.ok) {
+      const errorText = await response.text();
+      console.error('[可乐] STT API 错误:', response.status, errorText);
+      // 尝试解析 JSON 错误
+      try {
+        const errorJson = JSON.parse(errorText);
+        const errorMsg = errorJson.error?.message || errorJson.message || errorText;
+        throw new Error(errorMsg);
+      } catch (parseErr) {
+        // 如果不是 JSON 解析错误，而是 throw 的错误，重新抛出
+        if (parseErr.message && !parseErr.message.includes('JSON')) {
+          throw parseErr;
+        }
+        throw new Error(`HTTP ${response.status}: ${errorText.substring(0, 200)}`);
+      }
+    }
+
+    const result = await response.json();
+    console.log('[可乐] STT 响应:', result);
+    return result.text || '';
+  } catch (err) {
+    console.error('[可乐] STT 请求失败:', err);
+    throw err;
+  }
+}
+
+/**
+ * TTS: 文字转语音
+ * @param {string} text - 要合成的文字
+ * @param {Object} contact - 角色对象（用于获取角色独立音色）
+ * @param {Object} options - 选项
+ * @returns {Promise<Blob>} 音频 Blob
+ */
+export async function textToSpeech(text, contact = null, options = {}) {
+  const config = getVoiceApiConfig(contact);
+
+  if (!config.tts.url || !config.tts.key) {
+    throw new Error('请先配置语音合成 (TTS) API');
+  }
+
+  if (!text || !text.trim()) {
+    throw new Error('合成文字不能为空');
+  }
+
+  // 自动补全 URL 路径
+  let ttsUrl = config.tts.url.trim().replace(/\/+$/, '');
+  if (!ttsUrl.includes('/audio/speech')) {
+    ttsUrl = ttsUrl + '/audio/speech';
+  }
+
+  // 构建请求体
+  const model = (options.model || config.tts.model || '').trim();
+  const voice = (options.voice || config.tts.voice || '').trim();
+
+  // 检查必填字段
+  if (!model) {
+    throw new Error('请先配置 TTS 模型');
+  }
+  if (!voice) {
+    throw new Error('请先配置 TTS 音色');
+  }
+
+  // 检测是否是 Gemini TTS 模型
+  const isGeminiTTS = model.toLowerCase().includes('gemini') && model.toLowerCase().includes('tts');
+  // 检测是否是 GSVI 模型 (gsv2p.acgnai.top)
+  const isGSVI = model.toLowerCase().includes('gsvi');
+  // 检测是否是 MiniMax TTS API
+  const isMiniMax = ttsUrl.toLowerCase().includes('minimax') || ttsUrl.includes('/t2a_v2');
+
+  // MiniMax API 使用完全不同的格式
+  if (isMiniMax) {
+    // 修正 URL：MiniMax 使用 /v1/t2a_v2 而不是 /audio/speech
+    ttsUrl = ttsUrl.replace(/\/audio\/speech$/, '/t2a_v2');
+    if (!ttsUrl.includes('/t2a_v2')) {
+      ttsUrl = ttsUrl.replace(/\/+$/, '') + '/t2a_v2';
+    }
+
+    // 如果配置了代理 URL，使用代理（解决 CORS 问题）
+    if (config.tts.proxyUrl) {
+      const proxyBase = config.tts.proxyUrl.trim().replace(/\/+$/, '');
+      // 提取 MiniMax URL 的路径部分
+      const urlObj = new URL(ttsUrl);
+      ttsUrl = proxyBase + urlObj.pathname;
+      console.log('[可乐] MiniMax 使用代理:', ttsUrl);
+    }
+  }
+
+  // 构建请求体
+  let requestBody;
+
+  if (isMiniMax) {
+    // MiniMax API 格式
+    const speed = options.speed || config.tts.speed || 1;
+    const emotion = options.emotion || config.tts.emotion;
+
+    requestBody = {
+      model: model,
+      text: text.trim(),
+      stream: false,
+      voice_setting: {
+        voice_id: voice,
+        speed: speed,
+        vol: 1,
+        pitch: 0
+      },
+      audio_setting: {
+        sample_rate: 32000,
+        bitrate: 128000,
+        format: 'mp3',
+        channel: 1
+      }
+    };
+
+    // 添加情绪参数（只有有效值才添加）
+    if (emotion && emotion !== '默认') {
+      const emotionMap = {
+        '高兴': 'happy',
+        '悲伤': 'sad',
+        '愤怒': 'angry',
+        '害怕': 'fearful',
+        '厌恶': 'disgusted',
+        '惊讶': 'surprised',
+        '中性': 'calm',
+        '生动': 'fluent',
+        '低语': 'whisper'
+      };
+      // 只有在 emotionMap 中有对应值时才添加
+      const mappedEmotion = emotionMap[emotion];
+      if (mappedEmotion) {
+        requestBody.voice_setting.emotion = mappedEmotion;
+      }
+    }
+  } else {
+    requestBody = {
+      model: model,
+      voice: voice
+    };
+
+    // GSVI 模型只需要基本参数
+    if (isGSVI) {
+      requestBody.input = text.trim();
+      // GSVI API 不需要 language 和 emotion 参数
+    } else {
+      // OpenAI 标准格式使用 input
+      requestBody.input = text.trim();
+
+      // 非 Gemini TTS 时才添加额外参数
+      if (!isGeminiTTS) {
+        // 只有非默认语速才添加 speed 参数
+        const speed = options.speed || config.tts.speed || 1;
+        if (speed !== 1) {
+          requestBody.speed = speed;
+        }
+
+        // 扩展参数 (GPT-SoVITS 等支持)
+        const emotion = options.emotion || config.tts.emotion;
+        if (emotion && emotion !== '默认') {
+          requestBody.other_params = {
+            text_lang: '中英混合',
+            prompt_lang: '中文',
+            emotion: emotion
+          };
+        }
+      }
+    }
+  }
+
+  try {
+    const textContent = requestBody.input || requestBody.text || '';
+    console.log('[可乐] TTS 请求:', {
+      url: ttsUrl,
+      model: model,
+      voice: voice,
+      isGSVI: isGSVI,
+      isGeminiTTS: isGeminiTTS,
+      isMiniMax: isMiniMax,
+      textLength: textContent.length,
+      textFull: textContent  // 打印完整文本
+    });
+
+    const response = await fetch(ttsUrl, {
+      method: 'POST',
+      headers: {
+        'Content-Type': 'application/json',
+        'Accept': isMiniMax ? 'application/json' : 'audio/mpeg, audio/wav, audio/*',
+        'Authorization': `Bearer ${config.tts.key}`
+      },
+      body: JSON.stringify(requestBody)
+    });
+
+    if (!response.ok) {
+      const errorText = await response.text();
+      console.error('[可乐] TTS API 错误:');
+      console.error('  状态码:', response.status);
+      console.error('  响应内容:', errorText);
+      console.error('  请求URL:', ttsUrl);
+      console.error('  请求体:', JSON.stringify(requestBody, null, 2));
+
+      // 尝试解析 JSON 错误
+      try {
+        const errorJson = JSON.parse(errorText);
+        // MiniMax 错误格式: base_resp.status_msg
+        const errorMsg = errorJson.base_resp?.status_msg || errorJson.error?.message || errorJson.message || errorJson.error || errorText;
+        throw new Error(typeof errorMsg === 'string' ? errorMsg : JSON.stringify(errorMsg));
+      } catch (parseErr) {
+        if (parseErr.message && !parseErr.message.includes('JSON')) {
+          throw parseErr;
+        }
+        throw new Error(`HTTP ${response.status}: ${errorText.substring(0, 300)}`);
+      }
+    }
+
+    // MiniMax API 返回 JSON，需要特殊处理
+    if (isMiniMax) {
+      const jsonResp = await response.json();
+      console.log('[可乐] MiniMax TTS 响应:', {
+        status_code: jsonResp.base_resp?.status_code,
+        status_msg: jsonResp.base_resp?.status_msg,
+        audio_length: jsonResp.extra_info?.audio_length,
+        audio_format: jsonResp.extra_info?.audio_format
+      });
+
+      // 检查 MiniMax 错误
+      if (jsonResp.base_resp?.status_code !== 0) {
+        throw new Error('MiniMax TTS 错误: ' + (jsonResp.base_resp?.status_msg || '未知错误'));
+      }
+
+      if (!jsonResp.data?.audio) {
+        throw new Error('MiniMax TTS 未返回音频数据');
+      }
+
+      // 将 hex 编码的音频转换为 Blob
+      const hexAudio = jsonResp.data.audio;
+      const bytes = new Uint8Array(hexAudio.length / 2);
+      for (let i = 0; i < hexAudio.length; i += 2) {
+        bytes[i / 2] = parseInt(hexAudio.substr(i, 2), 16);
+      }
+
+      const audioFormat = jsonResp.extra_info?.audio_format || 'mp3';
+      const mimeType = `audio/${audioFormat}`;
+      return new Blob([bytes], { type: mimeType });
+    }
+
+    const audioBlob = await response.blob();
+    console.log('[可乐] TTS 响应:', {
+      音频大小: audioBlob.size,
+      类型: audioBlob.type,
+      响应头ContentType: response.headers.get('content-type')
+    });
+
+    // 先检查是否返回了错误的 JSON（有些 API 错误时返回 JSON）
+    const contentType = response.headers.get('content-type') || audioBlob.type;
+    if (contentType.includes('application/json') || contentType.includes('text/')) {
+      const text = await audioBlob.text();
+      console.error('[可乐] TTS 返回了文本而非音频:', text);
+      try {
+        const errJson = JSON.parse(text);
+        const errMsg = errJson.error?.message || errJson.message || errJson.error || JSON.stringify(errJson);
+        throw new Error('TTS 错误: ' + errMsg);
+      } catch (e) {
+        if (e.message.includes('TTS')) throw e;
+        throw new Error('TTS 返回了非音频数据: ' + text.substring(0, 100));
+      }
+    }
+
+    // 检查是否返回了有效的音频数据
+    if (audioBlob.size < 100) {
+      console.error('[可乐] TTS 返回的数据太小，可能不是有效音频');
+      throw new Error('TTS 返回的音频数据无效');
+    }
+
+    // 修复：如果 blob 类型为空或不是音频类型，手动指定 MIME 类型
+    // 某些 TTS API（如 GPT-SoVITS）返回的音频没有正确的 Content-Type
+    let finalBlob = audioBlob;
+    if (!audioBlob.type || audioBlob.type === '' || !audioBlob.type.startsWith('audio/')) {
+      // 尝试从 Content-Type 头获取类型，或使用默认的 audio/wav
+      let mimeType = 'audio/wav';
+      const headerType = response.headers.get('content-type');
+      if (headerType && headerType.startsWith('audio/')) {
+        mimeType = headerType.split(';')[0].trim();
+      } else if (headerType && headerType.includes('octet-stream')) {
+        // application/octet-stream 通常是 wav 格式
+        mimeType = 'audio/wav';
+      }
+
+      console.log('[可乐] TTS blob 类型为空，手动指定为:', mimeType);
+      const arrayBuffer = await audioBlob.arrayBuffer();
+      finalBlob = new Blob([arrayBuffer], { type: mimeType });
+    }
+
+    return finalBlob;
+  } catch (err) {
+    console.error('[可乐] TTS 请求失败:', err);
+    // 检查是否是网络错误
+    if (err.message?.includes('Failed to fetch') || err.message?.includes('NetworkError')) {
+      throw new Error('网络连接失败，请检查 API 地址是否正确，或尝试使用代理');
+    }
+    throw err;
+  }
+}
+
+/**
+ * 播放音频
+ * @param {Blob|string} audio - 音频 Blob 或 URL
+ * @returns {Promise<HTMLAudioElement>} Audio 元素
+ */
+export function playAudio(audio) {
+  return new Promise((resolve, reject) => {
+    const audioEl = new Audio();
+
+    if (audio instanceof Blob) {
+      audioEl.src = URL.createObjectURL(audio);
+    } else {
+      audioEl.src = audio;
+    }
+
+    audioEl.onended = () => {
+      if (audio instanceof Blob) {
+        URL.revokeObjectURL(audioEl.src);
+      }
+      resolve(audioEl);
+    };
+
+    audioEl.onerror = (err) => {
+      if (audio instanceof Blob) {
+        URL.revokeObjectURL(audioEl.src);
+      }
+      reject(err);
+    };
+
+    audioEl.play().catch(reject);
+  });
+}
+
+/**
+ * 录音类
+ */
+export class AudioRecorder {
+  constructor() {
+    this.mediaRecorder = null;
+    this.audioChunks = [];
+    this.stream = null;
+    this.isRecording = false;
+    this.mimeType = 'audio/webm';
+  }
+
+  /**
+   * 开始录音
+   * @returns {Promise<void>}
+   */
+  async start() {
+    if (this.isRecording) return;
+
+    try {
+      this.stream = await navigator.mediaDevices.getUserMedia({ audio: true });
+
+      // 选择最佳支持的音频格式
+      this.mimeType = getSupportedMimeType();
+      console.log('[可乐] 录音使用格式:', this.mimeType);
+
+      this.mediaRecorder = new MediaRecorder(this.stream, {
+        mimeType: this.mimeType
+      });
+      this.audioChunks = [];
+
+      this.mediaRecorder.ondataavailable = (e) => {
+        if (e.data.size > 0) {
+          this.audioChunks.push(e.data);
+        }
+      };
+
+      this.mediaRecorder.start(100); // 每100ms收集一次数据
+      this.isRecording = true;
+      console.log('[可乐] 开始录音');
+    } catch (err) {
+      console.error('[可乐] 无法获取麦克风权限:', err);
+      throw new Error('无法获取麦克风权限，请检查浏览器设置');
+    }
+  }
+
+  /**
+   * 停止录音
+   * @returns {Promise<Blob>} 录音数据
+   */
+  stop() {
+    return new Promise((resolve, reject) => {
+      if (!this.isRecording || !this.mediaRecorder) {
+        reject(new Error('没有正在进行的录音'));
+        return;
+      }
+
+      const mimeType = this.mimeType;
+
+      this.mediaRecorder.onstop = () => {
+        const audioBlob = new Blob(this.audioChunks, { type: mimeType });
+        this.cleanup();
+        console.log('[可乐] 录音结束，格式:', mimeType, '大小:', audioBlob.size);
+        resolve(audioBlob);
+      };
+
+      this.mediaRecorder.stop();
+      this.isRecording = false;
+    });
+  }
+
+  /**
+   * 取消录音
+   */
+  cancel() {
+    if (this.mediaRecorder && this.isRecording) {
+      this.mediaRecorder.stop();
+    }
+    this.cleanup();
+    this.isRecording = false;
+  }
+
+  /**
+   * 清理资源
+   */
+  cleanup() {
+    if (this.stream) {
+      this.stream.getTracks().forEach(track => track.stop());
+      this.stream = null;
+    }
+    this.mediaRecorder = null;
+    this.audioChunks = [];
+  }
+
+  /**
+   * 检查浏览器是否支持录音
+   * @returns {boolean}
+   */
+  static isSupported() {
+    return !!(navigator.mediaDevices && navigator.mediaDevices.getUserMedia);
+  }
+}
+
+/**
+ * 获取 MediaRecorder 支持的音频格式
+ */
+function getSupportedMimeType() {
+  const types = [
+    'audio/webm;codecs=opus',
+    'audio/webm',
+    'audio/ogg;codecs=opus',
+    'audio/ogg',
+    'audio/mp4',
+    'audio/mpeg'
+  ];
+
+  for (const type of types) {
+    if (MediaRecorder.isTypeSupported(type)) {
+      return type;
+    }
+  }
+  return 'audio/webm';
+}
+
+/**
+ * 测试 STT API
+ * @returns {Promise<boolean>}
+ */
+export async function testSttApi() {
+  const config = getVoiceApiConfig();
+
+  if (!config.stt.url || !config.stt.key) {
+    throw new Error('请先填写 STT API 地址和密钥');
+  }
+
+  console.log('[可乐] 开始 STT 测试...');
+  console.log('[可乐] STT 配置:', {
+    url: config.stt.url,
+    model: config.stt.model,
+    keyLength: config.stt.key?.length || 0
+  });
+
+  // 创建测试音频 (1.5秒，包含一些变化的音调模拟语音)
+  const audioContext = new (window.AudioContext || window.webkitAudioContext)();
+  const oscillator = audioContext.createOscillator();
+  const gainNode = audioContext.createGain();
+  const destination = audioContext.createMediaStreamDestination();
+
+  oscillator.connect(gainNode);
+  gainNode.connect(destination);
+
+  // 模拟语音的频率变化
+  oscillator.frequency.setValueAtTime(200, audioContext.currentTime);
+  oscillator.frequency.linearRampToValueAtTime(400, audioContext.currentTime + 0.5);
+  oscillator.frequency.linearRampToValueAtTime(300, audioContext.currentTime + 1);
+  oscillator.frequency.linearRampToValueAtTime(350, audioContext.currentTime + 1.5);
+
+  // 音量包络
+  gainNode.gain.setValueAtTime(0.3, audioContext.currentTime);
+  gainNode.gain.linearRampToValueAtTime(0.5, audioContext.currentTime + 0.3);
+  gainNode.gain.linearRampToValueAtTime(0.3, audioContext.currentTime + 1.2);
+  gainNode.gain.linearRampToValueAtTime(0, audioContext.currentTime + 1.5);
+
+  oscillator.start();
+
+  const mimeType = getSupportedMimeType();
+  console.log('[可乐] 录制音频格式:', mimeType);
+
+  const recorder = new MediaRecorder(destination.stream, { mimeType });
+  const chunks = [];
+
+  return new Promise((resolve, reject) => {
+    recorder.ondataavailable = e => {
+      if (e.data.size > 0) {
+        chunks.push(e.data);
+      }
+    };
+
+    recorder.onstop = async () => {
+      oscillator.stop();
+      audioContext.close();
+
+      const blob = new Blob(chunks, { type: mimeType });
+      console.log('[可乐] 测试音频大小:', blob.size, 'bytes');
+
+      if (blob.size < 100) {
+        reject(new Error('测试音频生成失败'));
+        return;
+      }
+
+      try {
+        // speechToText 会自动转换为 WAV 格式
+        const result = await speechToText(blob);
+        console.log('[可乐] STT 测试结果:', result);
+        resolve(true);
+      } catch (err) {
+        reject(err);
+      }
+    };
+
+    recorder.start(100);
+    // 录制 1.5 秒
+    setTimeout(() => recorder.stop(), 1500);
+  });
+}
+
+/**
+ * 测试 TTS API
+ * @returns {Promise<Blob>}
+ */
+export async function testTtsApi() {
+  const config = getVoiceApiConfig();
+
+  if (!config.tts.url || !config.tts.key) {
+    throw new Error('请先填写 TTS API 地址和密钥');
+  }
+
+  return await textToSpeech('测试语音合成');
+}

voice-call.js

@@ -106,6 +106,11 @@ function showCallPage() {
   if (chatEl) {
     chatEl.classList.add('hidden');
   }
+  // 隐藏输入框
+  const inputAreaEl = document.getElementById('wechat-voice-call-input-area');
+  if (inputAreaEl) {
+    inputAreaEl.classList.add('hidden');
+  }
   const messagesEl = document.getElementById('wechat-voice-call-messages');
   if (messagesEl) {
     messagesEl.innerHTML = '';
@@ -243,6 +248,11 @@ function onCallConnected() {
   if (chatEl) {
     chatEl.classList.remove('hidden');
   }
+  // 显示输入框
+  const inputAreaEl = document.getElementById('wechat-voice-call-input-area');
+  if (inputAreaEl) {
+    inputAreaEl.classList.remove('hidden');
+  }
 
   // 切换到通话中按钮（隐藏来电按钮，显示通话控制按钮）
   const incomingActionsEl = document.getElementById('wechat-voice-call-incoming-actions');