添加了 TTS 相关服务并更新了设置

src/main/services/MsTTSIpcHandler.ts

@@ -16,32 +16,35 @@ export function registerMsTTSIpcHandlers(): void {
   )
 
   // 流式合成语音
-  ipcMain.handle(IpcChannel.MsTTS_SynthesizeStream, async (event, requestId: string, text: string, voice: string, outputFormat: string) => {
+  ipcMain.handle(
-    const window = BrowserWindow.fromWebContents(event.sender)
+    IpcChannel.MsTTS_SynthesizeStream,
-    if (!window) return
+    async (event, requestId: string, text: string, voice: string, outputFormat: string) => {
+      const window = BrowserWindow.fromWebContents(event.sender)
+      if (!window) return
 
-    try {
+      try {
-      await MsTTSService.synthesizeStream(
+        await MsTTSService.synthesizeStream(
-        text,
+          text,
-        voice,
+          voice,
-        outputFormat,
+          outputFormat,
-        (chunk: Uint8Array) => {
+          (chunk: Uint8Array) => {
-          // 发送音频数据块
+            // 发送音频数据块
-          if (!window.isDestroyed()) {
+            if (!window.isDestroyed()) {
-            window.webContents.send(IpcChannel.MsTTS_StreamData, requestId, chunk)
+              window.webContents.send(IpcChannel.MsTTS_StreamData, requestId, chunk)
+            }
+          },
+          () => {
+            // 发送流结束信号
+            if (!window.isDestroyed()) {
+              window.webContents.send(IpcChannel.MsTTS_StreamEnd, requestId)
+            }
           }
-        },
+        )
-        () => {
+        return { success: true }
-          // 发送流结束信号
+      } catch (error) {
-          if (!window.isDestroyed()) {
+        console.error('流式TTS合成失败:', error)
-            window.webContents.send(IpcChannel.MsTTS_StreamEnd, requestId)
+        return { success: false, error: error instanceof Error ? error.message : String(error) }
-          }
+      }
-        }
-      )
-      return { success: true }
-    } catch (error) {
-      console.error('流式TTS合成失败:', error)
-      return { success: false, error: error instanceof Error ? error.message : String(error) }
     }
-  })
+  )
 }

src/main/services/MsTTSService.ts

@@ -1,10 +1,10 @@
 import fs from 'node:fs'
 import path from 'node:path'
 
+import { MsEdgeTTS, OUTPUT_FORMAT } from 'edge-tts-node' // 新版支持流式的TTS库
 import { app } from 'electron'
 import log from 'electron-log'
 import { EdgeTTS } from 'node-edge-tts' // 旧版TTS库
-import { MsEdgeTTS, OUTPUT_FORMAT } from 'edge-tts-node' // 新版支持流式的TTS库
 
 // --- START OF HARDCODED VOICE LIST ---
 // WARNING: This list is static and may become outdated.

src/renderer/src/components/ASRButton.tsx

@@ -33,7 +33,7 @@ const ASRButton: FC<Props> = ({ onTranscribed, disabled = false, style }) => {
       try {
         // 添加事件监听器，监听服务器发送的stopped消息
         const originalCallback = ASRService.resultCallback
-        const stopCallback = (text: string, isFinal?: boolean) => {
+        const stopCallback = (text: string) => {
           // 如果是空字符串，只重置状态，不调用原始回调
           if (text === '') {
             setIsProcessing(false)

src/renderer/src/components/DraggableVoiceCallWindow.tsx

@@ -289,7 +289,9 @@ const DraggableVoiceCallWindow: React.FC<Props> = ({
         }
         await VoiceCallService.startCall({
           onTranscript: setTranscript,
-          onResponse: () => { /* 响应在聊天界面处理 */ },
+          onResponse: () => {
+            /* 响应在聊天界面处理 */
+          },
           onListeningStateChange: setIsListening,
           onSpeakingStateChange: setIsSpeaking
         })
@@ -402,7 +404,7 @@ const DraggableVoiceCallWindow: React.FC<Props> = ({
           await VoiceCallService.cancelRecording()
           setTranscript('')
         } catch (error) {
-          console.error('取消录音出错:', error);
+          console.error('取消录音出错:', error)
         } finally {
           setTimeout(() => setIsProcessing(false), 500)
         }
@@ -412,23 +414,31 @@ const DraggableVoiceCallWindow: React.FC<Props> = ({
   )
   // --- 语音通话控制函数结束 ---
 
-
   // --- 快捷键相关函数 ---
   const getKeyDisplayName = (keyCode: string) => {
     const keyMap: Record<string, string> = {
-      Space: '空格键', Enter: '回车键', ShiftLeft: '左Shift键', ShiftRight: '右Shift键',
+      Space: '空格键',
-      ControlLeft: '左Ctrl键', ControlRight: '右Ctrl键', AltLeft: '左Alt键', AltRight: '右Alt键'
+      Enter: '回车键',
+      ShiftLeft: '左Shift键',
+      ShiftRight: '右Shift键',
+      ControlLeft: '左Ctrl键',
+      ControlRight: '右Ctrl键',
+      AltLeft: '左Alt键',
+      AltRight: '右Alt键'
     }
     return keyMap[keyCode] || keyCode
   }
 
-  const handleShortcutKeyChange = useCallback((e: KeyboardEvent) => {
+  const handleShortcutKeyChange = useCallback(
-    e.preventDefault()
+    (e: KeyboardEvent) => {
-    if (isRecordingShortcut) {
+      e.preventDefault()
-      setTempShortcutKey(e.code)
+      if (isRecordingShortcut) {
-      setIsRecordingShortcut(false)
+        setTempShortcutKey(e.code)
-    }
+        setIsRecordingShortcut(false)
-  }, [isRecordingShortcut])
+      }
+    },
+    [isRecordingShortcut]
+  )
 
   const saveShortcutKey = useCallback(() => {
     setShortcutKey(tempShortcutKey)
@@ -437,31 +447,42 @@ const DraggableVoiceCallWindow: React.FC<Props> = ({
   }, [tempShortcutKey])
 
   // 现在可以安全地使用 handleRecordStart/End
-  const handleKeyDown = useCallback((e: KeyboardEvent) => {
+  const handleKeyDown = useCallback(
-    if (isRecordingShortcut) {
+    (e: KeyboardEvent) => {
-      handleShortcutKeyChange(e)
+      if (isRecordingShortcut) {
-      return
+        handleShortcutKeyChange(e)
-    }
+        return
-    if (e.code === shortcutKey && !isProcessing && !isPaused && visible && !isShortcutPressed) {
+      }
-      e.preventDefault()
+      if (e.code === shortcutKey && !isProcessing && !isPaused && visible && !isShortcutPressed) {
-      setIsShortcutPressed(true)
+        e.preventDefault()
-      const mockEvent = new MouseEvent('mousedown') as unknown as React.MouseEvent // 类型断言
+        setIsShortcutPressed(true)
-      handleRecordStart(mockEvent) // 现在 handleRecordStart 已经定义
+        const mockEvent = new MouseEvent('mousedown') as unknown as React.MouseEvent // 类型断言
-    }
+        handleRecordStart(mockEvent) // 现在 handleRecordStart 已经定义
-  }, [
+      }
-    shortcutKey, isProcessing, isPaused, visible, isShortcutPressed,
+    },
-    handleRecordStart, // 依赖项
+    [
-    isRecordingShortcut, handleShortcutKeyChange
+      shortcutKey,
-  ])
+      isProcessing,
+      isPaused,
+      visible,
+      isShortcutPressed,
+      handleRecordStart, // 依赖项
+      isRecordingShortcut,
+      handleShortcutKeyChange
+    ]
+  )
 
-  const handleKeyUp = useCallback((e: KeyboardEvent) => {
+  const handleKeyUp = useCallback(
-    if (e.code === shortcutKey && isShortcutPressed && visible) {
+    (e: KeyboardEvent) => {
-      e.preventDefault()
+      if (e.code === shortcutKey && isShortcutPressed && visible) {
-      setIsShortcutPressed(false)
+        e.preventDefault()
-      const mockEvent = new MouseEvent('mouseup') as unknown as React.MouseEvent // 类型断言
+        setIsShortcutPressed(false)
-      handleRecordEnd(mockEvent) // 现在 handleRecordEnd 已经定义
+        const mockEvent = new MouseEvent('mouseup') as unknown as React.MouseEvent // 类型断言
-    }
+        handleRecordEnd(mockEvent) // 现在 handleRecordEnd 已经定义
-  }, [shortcutKey, isShortcutPressed, visible, handleRecordEnd]) // 依赖项
+      }
+    },
+    [shortcutKey, isShortcutPressed, visible, handleRecordEnd]
+  ) // 依赖项
 
   useEffect(() => {
     const savedShortcut = localStorage.getItem('voiceCallShortcutKey')
@@ -483,7 +504,6 @@ const DraggableVoiceCallWindow: React.FC<Props> = ({
   }, [visible, handleKeyDown, handleKeyUp])
   // --- 快捷键相关函数结束 ---
 
-
   // 如果不可见，直接返回 null
   if (!visible) return null
 
@@ -511,10 +531,14 @@ const DraggableVoiceCallWindow: React.FC<Props> = ({
 
       <Content>
         {isSettingsVisible && (
-          <SettingsPanel> {/* 使用 styled-component */}
+          <SettingsPanel>
+            {' '}
+            {/* 使用 styled-component */}
             <SettingsTitle>{t('voice_call.shortcut_key_setting')}</SettingsTitle> {/* 使用 styled-component */}
             <Space>
-              <ShortcutKeyButton onClick={() => setIsRecordingShortcut(true)}> {/* 使用 styled-component */}
+              <ShortcutKeyButton onClick={() => setIsRecordingShortcut(true)}>
+                {' '}
+                {/* 使用 styled-component */}
                 {isRecordingShortcut ? t('voice_call.press_any_key') : getKeyDisplayName(tempShortcutKey)}
               </ShortcutKeyButton>
               <Button type="primary" onClick={saveShortcutKey}>
@@ -522,7 +546,9 @@ const DraggableVoiceCallWindow: React.FC<Props> = ({
               </Button>
               <Button onClick={() => setIsSettingsVisible(false)}>{t('voice_call.cancel')}</Button>
             </Space>
-            <SettingsTip> {/* 使用 styled-component */}
+            <SettingsTip>
+              {' '}
+              {/* 使用 styled-component */}
               {t('voice_call.shortcut_key_tip')}
             </SettingsTip>
           </SettingsPanel>
@@ -579,4 +605,4 @@ const DraggableVoiceCallWindow: React.FC<Props> = ({
   )
 }
 
-export default DraggableVoiceCallWindow
+export default DraggableVoiceCallWindow

src/renderer/src/i18n/locales/en-us.json

@@ -15,7 +15,12 @@
       "initialization_failed": "Failed to initialize voice call",
       "error": "Voice call error",
       "initializing": "Initializing voice call...",
-      "ready": "Voice call ready"
+      "ready": "Voice call ready",
+      "shortcut_key_setting": "[to be translated]:语音识别快捷键设置",
+      "press_any_key": "[to be translated]:请按任意键...",
+      "save": "[to be translated]:保存",
+      "cancel": "[to be translated]:取消",
+      "shortcut_key_tip": "[to be translated]:按下此快捷键开始录音，松开快捷键结束录音并发送"
     },
     "agents": {
       "add.button": "Add to Assistant",

src/renderer/src/i18n/locales/ja-jp.json

@@ -1421,8 +1421,7 @@
         "mstts.voice": "[to be translated]:免费在线 TTS音色",
         "mstts.output_format": "[to be translated]:输出格式",
         "mstts.info": "[to be translated]:免费在线TTS服务不需要API密钥，完全免费使用。",
-        "error.no_mstts_voice": "[to be translated]:未设置免费在线 TTS音色",
+        "error.no_mstts_voice": "[to be translated]:未设置免费在线 TTS音色"
-        "filter.emojis": "[to be translated]:过滤表情符号"
       },
       "asr": {
         "title": "音声認識",
@@ -1544,7 +1543,12 @@
       "initialization_failed": "[to be translated]:初始化语音通话失败",
       "error": "[to be translated]:语音通话出错",
       "initializing": "[to be translated]:正在初始化语音通话...",
-      "ready": "[to be translated]:语音通话已就绪"
+      "ready": "[to be translated]:语音通话已就绪",
+      "shortcut_key_setting": "[to be translated]:语音识别快捷键设置",
+      "press_any_key": "[to be translated]:请按任意键...",
+      "save": "[to be translated]:保存",
+      "cancel": "[to be translated]:取消",
+      "shortcut_key_tip": "[to be translated]:按下此快捷键开始录音，松开快捷键结束录音并发送"
     }
   }
 }

src/renderer/src/i18n/locales/ru-ru.json

@@ -1421,8 +1421,7 @@
         "mstts.voice": "[to be translated]:免费在线 TTS音色",
         "mstts.output_format": "[to be translated]:输出格式",
         "mstts.info": "[to be translated]:免费在线TTS服务不需要API密钥，完全免费使用。",
-        "error.no_mstts_voice": "[to be translated]:未设置免费在线 TTS音色",
+        "error.no_mstts_voice": "[to be translated]:未设置免费在线 TTS音色"
-        "filter.emojis": "[to be translated]:过滤表情符号"
       },
       "voice": {
         "title": "[to be translated]:语音功能",
@@ -1544,7 +1543,12 @@
       "initialization_failed": "[to be translated]:初始化语音通话失败",
       "error": "[to be translated]:语音通话出错",
       "initializing": "[to be translated]:正在初始化语音通话...",
-      "ready": "[to be translated]:语音通话已就绪"
+      "ready": "[to be translated]:语音通话已就绪",
+      "shortcut_key_setting": "[to be translated]:语音识别快捷键设置",
+      "press_any_key": "[to be translated]:请按任意键...",
+      "save": "[to be translated]:保存",
+      "cancel": "[to be translated]:取消",
+      "shortcut_key_tip": "[to be translated]:按下此快捷键开始录音，松开快捷键结束录音并发送"
     }
   }
 }

src/renderer/src/i18n/locales/zh-tw.json

@@ -1421,8 +1421,7 @@
         "mstts.voice": "[to be translated]:免费在线 TTS音色",
         "mstts.output_format": "[to be translated]:输出格式",
         "mstts.info": "[to be translated]:免费在线TTS服务不需要API密钥，完全免费使用。",
-        "error.no_mstts_voice": "[to be translated]:未设置免费在线 TTS音色",
+        "error.no_mstts_voice": "[to be translated]:未设置免费在线 TTS音色"
-        "filter.emojis": "[to be translated]:过滤表情符号"
       },
       "voice": {
         "title": "[to be translated]:语音功能",
@@ -1544,7 +1543,12 @@
       "initialization_failed": "[to be translated]:初始化语音通话失败",
       "error": "[to be translated]:语音通话出错",
       "initializing": "[to be translated]:正在初始化语音通话...",
-      "ready": "[to be translated]:语音通话已就绪"
+      "ready": "[to be translated]:语音通话已就绪",
+      "shortcut_key_setting": "[to be translated]:语音识别快捷键设置",
+      "press_any_key": "[to be translated]:请按任意键...",
+      "save": "[to be translated]:保存",
+      "cancel": "[to be translated]:取消",
+      "shortcut_key_tip": "[to be translated]:按下此快捷键开始录音，松开快捷键结束录音并发送"
     }
   }
 }

src/renderer/src/pages/home/Inputbar/Inputbar.tsx

@@ -77,7 +77,8 @@ let _files: FileType[] = []
 
 const Inputbar: FC<Props> = ({ assistant: _assistant, setActiveTopic, topic }) => {
   const [text, setText] = useState(_text)
-  const [asrCurrentText, setAsrCurrentText] = useState('')
+  // 用于存储语音识别的中间结果，不直接显示在输入框中
+  const [, setAsrCurrentText] = useState('')
   const [inputFocus, setInputFocus] = useState(false)
   const { assistant, addTopic, model, setModel, updateAssistant } = useAssistant(_assistant.id)
   const {

src/renderer/src/pages/home/Messages/Message.tsx

@@ -155,13 +155,7 @@ const MessageItem: FC<Props> = ({
   useEffect(() => {
     // 如果是最后一条助手消息，且消息状态为成功，且不是正在生成中，且TTS已启用
     // 注意：只有在语音通话窗口打开时才自动播放TTS
-    if (
+    if (isLastMessage && isAssistantMessage && message.status === 'success' && !generating && ttsEnabled) {
-      isLastMessage &&
-      isAssistantMessage &&
-      message.status === 'success' &&
-      !generating &&
-      ttsEnabled
-    ) {
       // 如果语音通话窗口没有打开，则不自动播放TTS
       if (!isVoiceCallActive) {
         console.log('不自动播放TTS，因为语音通话窗口没有打开:', isVoiceCallActive)

src/renderer/src/services/ASRService.ts

@@ -170,7 +170,8 @@ class ASRService {
               // 直接调用回调函数
               this.resultCallback(data.data.text, true)
               window.message.success({ content: i18n.t('settings.asr.success'), key: 'asr-processing' })
-            } else if (this.isRecording) { // 只在录音中才处理中间结果
+            } else if (this.isRecording) {
+              // 只在录音中才处理中间结果
               // 非最终结果，也调用回调，但标记为非最终
               console.log('[ASRService] 收到中间结果，调用回调函数，文本:', data.data.text)
               this.resultCallback(data.data.text, false)

src/renderer/src/services/tts/AudioStreamProcessor.ts

@@ -6,10 +6,10 @@ export class AudioStreamProcessor {
   private audioContext: AudioContext | null = null
   private audioQueue: Uint8Array[] = []
   private isProcessing: boolean = false
-  
+
   // 回调函数
   public onAudioBuffer: ((buffer: AudioBuffer) => void) | null = null
-  
+
   /**
    * 初始化音频处理器
    */
@@ -19,7 +19,7 @@ export class AudioStreamProcessor {
     this.audioQueue = []
     this.isProcessing = false
   }
-  
+
   /**
    * 处理音频数据块
    * @param chunk 音频数据块
@@ -28,16 +28,16 @@ export class AudioStreamProcessor {
     if (!this.audioContext) {
       throw new Error('AudioStreamProcessor not initialized')
     }
-    
+
     // 将数据块添加到队列
     this.audioQueue.push(chunk)
-    
+
     // 如果没有正在处理，开始处理
     if (!this.isProcessing) {
       this.processQueue()
     }
   }
-  
+
   /**
    * 处理队列中的音频数据
    */
@@ -46,20 +46,18 @@ export class AudioStreamProcessor {
       this.isProcessing = false
       return
     }
-    
+
     this.isProcessing = true
-    
+
     // 获取队列中的第一个数据块
     const chunk = this.audioQueue.shift()!
-    
+
     try {
       // 解码音频数据
       // 将SharedArrayBuffer转换为ArrayBuffer
-      const arrayBuffer = chunk.buffer instanceof SharedArrayBuffer 
+      const arrayBuffer = chunk.buffer instanceof SharedArrayBuffer ? new Uint8Array(chunk.buffer).buffer : chunk.buffer
-        ? new Uint8Array(chunk.buffer).buffer 
+      const audioBuffer = await this.audioContext.decodeAudioData(arrayBuffer as ArrayBuffer)
-        : chunk.buffer
+
-      const audioBuffer = await this.audioContext.decodeAudioData(arrayBuffer)
-      
       // 调用回调函数
       if (this.onAudioBuffer) {
         this.onAudioBuffer(audioBuffer)
@@ -67,20 +65,20 @@ export class AudioStreamProcessor {
     } catch (error) {
       console.error('解码音频数据失败:', error)
     }
-    
+
     // 继续处理队列中的下一个数据块
     this.processQueue()
   }
-  
+
   /**
    * 完成处理
    */
   public async finish(): Promise<void> {
     // 等待队列处理完成
     while (this.audioQueue.length > 0) {
-      await new Promise(resolve => setTimeout(resolve, 100))
+      await new Promise((resolve) => setTimeout(resolve, 100))
     }
-    
+
     // 关闭音频上下文
     if (this.audioContext) {
       await this.audioContext.close()

src/renderer/src/services/tts/TTSService.ts

@@ -63,13 +63,13 @@ export class TTSService {
         // 只有在使用EdgeTTS且标记为正在播放时才检查
         if (this.isPlaying && this.playingServiceType === 'edge') {
           // 检查是否还在播放
-          const isSpeaking = window.speechSynthesis.speaking;
+          const isSpeaking = window.speechSynthesis.speaking
           if (!isSpeaking) {
             console.log('检测到speechSynthesis不再播放，更新状态')
-            this.updatePlayingState(false);
+            this.updatePlayingState(false)
           }
         }
-      }, 500); // 每500毫秒检查一次
+      }, 500) // 每500毫秒检查一次
     }
   }
 
@@ -109,7 +109,7 @@ export class TTSService {
   private updatePlayingState(isPlaying: boolean): void {
     // 只有状态变化时才更新和触发事件
     if (this.isPlaying !== isPlaying) {
-      this.isPlaying = isPlaying;
+      this.isPlaying = isPlaying
       console.log(`TTS播放状态更新: ${isPlaying ? '开始播放' : '停止播放'}`)
 
       // 触发自定义事件，通知其他组件TTS状态变化
@@ -118,11 +118,11 @@ export class TTSService {
 
       // 如果停止播放，清除服务类型
       if (!isPlaying) {
-        this.playingServiceType = null;
+        this.playingServiceType = null
 
         // 确保Web Speech API也停止
         if ('speechSynthesis' in window) {
-          window.speechSynthesis.cancel();
+          window.speechSynthesis.cancel()
         }
       }
     }
@@ -148,7 +148,7 @@ export class TTSService {
       if (this.isPlaying) {
         this.stop()
         // 添加短暂延迟，确保上一个播放完全停止
-        await new Promise(resolve => setTimeout(resolve, 100))
+        await new Promise((resolve) => setTimeout(resolve, 100))
       }
 
       // 确保文本不为空

src/renderer/src/services/tts/TTSServiceInterface.ts

@@ -19,5 +19,11 @@ export interface TTSServiceInterface {
    * @param onError 错误回调
    * @returns 返回请求ID
    */
-  synthesizeStream?(text: string, onStart: () => void, onData: (audioChunk: AudioBuffer) => void, onEnd: () => void, onError: (error: Error) => void): Promise<string>
+  synthesizeStream?(
+    text: string,
+    onStart: () => void,
+    onData: (audioChunk: AudioBuffer) => void,
+    onEnd: () => void,
+    onError: (error: Error) => void
+  ): Promise<string>
 }