From a42b751ed4a5d1ee0622bedc65ef1bd5b5b9cdff Mon Sep 17 00:00:00 2001 From: fdp <1286779656@qq.com> Date: Mon, 11 Aug 2025 20:05:23 +0800 Subject: [PATCH] =?UTF-8?q?=E9=9F=B3=E9=A2=91=E6=95=B0=E6=8D=AE=E6=8A=9B?= =?UTF-8?q?=E5=9B=9Eflutter=E5=B1=82?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- lib/data/services/asr_service.dart | 8 +- .../speech_impl/azure_asr_service.dart | 72 ++++++++++- .../speech_impl/volcano_asr_api_service.dart | 8 +- .../speech_impl/volcano_asr_service.dart | 8 +- .../speech_impl/xunfei_asr_service.dart | 8 +- .../meeting_record_controller.dart | 17 ++- .../controllers/translation_controller.dart | 118 +++++++++++++----- .../translation/views/translation_view.dart | 17 ++- .../azure_speech/AzureAsrHelper.kt | 68 +++++----- .../azure_speech/AzureAsrToAsr.kt | 1 + .../azure_speech/AzureSpeechPlugin.kt | 59 ++++++++- .../azure_speech/AzureTtsHelper.kt | 2 +- .../azure_speech/tools/SimpleAudioPlayer.kt | 2 +- .../azure_speech/tools/SimpleAudioReceiver.kt | 31 ++++- 14 files changed, 338 insertions(+), 81 deletions(-) diff --git a/lib/data/services/asr_service.dart b/lib/data/services/asr_service.dart index 1fba23f52..827194a1a 100644 --- a/lib/data/services/asr_service.dart +++ b/lib/data/services/asr_service.dart @@ -41,7 +41,13 @@ abstract class AsrService { bool isContinuousRecognitionActive(); /// 开始录音 - Future enableRecord(String filePath); + /// + /// [filePath] 录音文件路径 + /// [acceptAudioData] 是否接受音频数据回调，默认为 false + Future enableRecord(String filePath, {bool acceptAudioData = false}); + + /// 获取音频数据流（如果支持） + Stream? getAudioDataStream() => null; /// 暂停录音 Future pauseRecord(); diff --git a/lib/data/services/speech_impl/azure_asr_service.dart b/lib/data/services/speech_impl/azure_asr_service.dart index 0009cd758..7c34a204d 100644 --- a/lib/data/services/speech_impl/azure_asr_service.dart +++ b/lib/data/services/speech_impl/azure_asr_service.dart @@ -21,6 +21,10 @@ class AzureAsrService extends GetxService implements AsrService { static const MethodChannel _channel = MethodChannel('azure_speech/asr'); static const EventChannel _eventChannel = EventChannel('azure_speech/asr_events'); + // 音频数据事件通道 + static const EventChannel _audioDataEventChannel = + EventChannel('azure_speech/audio_data_events'); + // final GetStorage _storage = GetStorage(); bool _isInitialized = false; late final String _subscriptionKey; @@ -43,6 +47,11 @@ class AzureAsrService extends GetxService implements AsrService { String _latestDetectedLanguage = ''; String get latestDetectedLanguage => _latestDetectedLanguage; +// 音频数据事件订阅 + StreamSubscription? _audioDataEventSubscription; + // 音频数据流控制器 + StreamController? _audioDataStreamController; + // 当前音频源类型 AudioSourceType _audioSourceType = AudioSourceType.microphone; @@ -72,6 +81,19 @@ class AzureAsrService extends GetxService implements AsrService { }, onError: _handleRecognitionError); } + /// 设置音频数据事件通道 + void _setupAudioDataEventChannel() { + _audioDataEventSubscription?.cancel(); + _audioDataEventSubscription = + _audioDataEventChannel.receiveBroadcastStream().listen((event) { + if (event is Map) { + _handleAudioDataEvent(event); + } + }, onError: (error) { + Logger.error('音频数据事件流错误: ${error.toString()}'); + }); + } + @override Future initialize({ List? supportedLanguages, @@ -220,6 +242,41 @@ class AzureAsrService extends GetxService implements AsrService { return _isContinuousRecognitionActive; } + /// 处理音频数据事件 + void _handleAudioDataEvent(dynamic event) { + if (event is! Map) return; + + final Map eventMap = event; + final String eventType = eventMap['type'] as String? ?? ''; + + switch (eventType) { + case 'audioData': + final Uint8List data = eventMap['data'] as Uint8List? ?? Uint8List(0); + final double timestamp = + (eventMap['timestamp'] as num?)?.toDouble() ?? 0.0; + final int size = eventMap['size'] as int? ?? 0; + print('接收到音频数据: 大小=${size}字节, 时间戳=${timestamp}'); + Logger.debug('接收到音频数据: 大小=${size}字节, 时间戳=${timestamp}'); + + // 发送到音频数据流 + _audioDataStreamController?.add(data); + + // 也可以通过识别事件流发送 + _eventStreamController?.add(RecognitionEvent( + type: RecognitionEventType.onAudio, + audio: data, + )); + break; + } + } + + /// 获取音频数据流 + Stream? getAudioDataStream() { + print("getAudioDataStream"); + _audioDataStreamController ??= StreamController.broadcast(); + return _audioDataStreamController?.stream; + } + /// 处理来自原生端的识别事件 void _handleRecognitionEvent(dynamic event) { if (event is! Map || _eventStreamController == null) return; @@ -336,6 +393,14 @@ class AzureAsrService extends GetxService implements AsrService { await _eventSubscription?.cancel(); _eventSubscription = null; + // 取消音频数据事件订阅 + await _audioDataEventSubscription?.cancel(); + _audioDataEventSubscription = null; + + // 关闭音频数据流 + await _audioDataStreamController?.close(); + _audioDataStreamController = null; + // 通知原生端释放资源 await _channel.invokeMethod('dispose'); _isInitialized = false; @@ -376,11 +441,16 @@ class AzureAsrService extends GetxService implements AsrService { } @override - Future enableRecord(String filePath) async { + Future enableRecord(String filePath, + {bool acceptAudioData = false}) async { try { final bool result = await _channel.invokeMethod('enableRecord', { 'filePath': filePath, + 'acceptAudioData': acceptAudioData, // 新增参数 }); + if (acceptAudioData) { + _setupAudioDataEventChannel(); + } return result; } catch (e) { diff --git a/lib/data/services/speech_impl/volcano_asr_api_service.dart b/lib/data/services/speech_impl/volcano_asr_api_service.dart index 126aea878..87e0863d8 100644 --- a/lib/data/services/speech_impl/volcano_asr_api_service.dart +++ b/lib/data/services/speech_impl/volcano_asr_api_service.dart @@ -813,7 +813,7 @@ class VolcanoAsrApiService implements AsrService { } @override - Future enableRecord(String filePath) { + Future enableRecord(String filePath, {bool acceptAudioData = false}) { // TODO: implement enableRecord throw UnimplementedError(); } @@ -877,4 +877,10 @@ class VolcanoAsrApiService implements AsrService { // TODO: implement resumeRecord throw UnimplementedError(); } + + @override + Stream? getAudioDataStream() { + // TODO: implement getAudioDataStream + throw UnimplementedError(); + } } diff --git a/lib/data/services/speech_impl/volcano_asr_service.dart b/lib/data/services/speech_impl/volcano_asr_service.dart index 95d9c1afc..6d090d642 100644 --- a/lib/data/services/speech_impl/volcano_asr_service.dart +++ b/lib/data/services/speech_impl/volcano_asr_service.dart @@ -348,7 +348,7 @@ class VolcanoAsrService extends GetxService implements AsrService { } @override - Future enableRecord(String filePath) { + Future enableRecord(String filePath, {bool acceptAudioData = false}) { // TODO: implement enableRecord throw UnimplementedError(); } @@ -412,4 +412,10 @@ class VolcanoAsrService extends GetxService implements AsrService { // TODO: implement resumeRecord throw UnimplementedError(); } + + @override + Stream? getAudioDataStream() { + // TODO: implement getAudioDataStream + throw UnimplementedError(); + } } diff --git a/lib/data/services/speech_impl/xunfei_asr_service.dart b/lib/data/services/speech_impl/xunfei_asr_service.dart index ea8a52251..29daf7015 100644 --- a/lib/data/services/speech_impl/xunfei_asr_service.dart +++ b/lib/data/services/speech_impl/xunfei_asr_service.dart @@ -272,7 +272,7 @@ class XunfeiAsrService extends GetxService implements AsrService { } @override - Future enableRecord(String filePath) { + Future enableRecord(String filePath, {bool acceptAudioData = false}) { // TODO: implement enableRecord throw UnimplementedError(); } @@ -336,4 +336,10 @@ class XunfeiAsrService extends GetxService implements AsrService { // TODO: implement resumeRecord throw UnimplementedError(); } + + @override + Stream? getAudioDataStream() { + // TODO: implement getAudioDataStream + throw UnimplementedError(); + } } diff --git a/lib/modules/meeting/controllers/meeting_record_controller.dart b/lib/modules/meeting/controllers/meeting_record_controller.dart index 71ce8796e..eb177d910 100644 --- a/lib/modules/meeting/controllers/meeting_record_controller.dart +++ b/lib/modules/meeting/controllers/meeting_record_controller.dart @@ -60,6 +60,7 @@ class MeetingRecordController extends GetxController bool _audioSourceType = false; late TabController tabController; + StreamSubscription? _audioDataSubscription; // 添加波形放大倍数变量 final double waveAmplifyFactor = 100.0; // 增加波动幅度 final RxDouble ursorPosition = 0.0.obs; // 当前播放位置 @@ -255,7 +256,21 @@ class MeetingRecordController extends GetxController final fullFileName = "${fileName.value}_$formattedTime"; // Start audio recording - await _asrService.enableRecord("${dir.path}/$fullFileName.wav"); + await _asrService.enableRecord("${dir.path}/$fullFileName.wav", + acceptAudioData: true); + + // 监听音频数据流 + _audioDataSubscription = + _asrService.getAudioDataStream()?.listen((audioData) { + // 处理音频数据，例如： + // 1. 实时音频可视化 + // 2. 音频质量检测 + // 3. 发送到其他服务 + print('接收到音频数据: ${audioData.length} 字节'); + + // 处理音频数据的逻辑 + _processAudioData(audioData); + }); isRecording.value = true; // Update state diff --git a/lib/modules/translation/controllers/translation_controller.dart b/lib/modules/translation/controllers/translation_controller.dart index f4c81e9fd..52a2ad2c2 100644 --- a/lib/modules/translation/controllers/translation_controller.dart +++ b/lib/modules/translation/controllers/translation_controller.dart @@ -65,6 +65,8 @@ class TranslationController extends GetxController { Timer? _recordTimer; //秒 int _seconds = 0; + //暂停状态记录计时器是否暂停 + bool _isTimerPaused = false; // 存储相关配置 static const String _historyKey = 'translation_history'; static const String _sourceLanguageKey = 'translation_source_language'; @@ -92,6 +94,8 @@ class TranslationController extends GetxController { final isTranslating = false.obs; final isTtsEnabled = true.obs; final isRecording = false.obs; + final lasyIsRecording = false.obs; + final hasRecordPermission = false.obs; // 语言相关 final sourceLanguage = '中文(简体)'.obs; final targetLanguage = '英语'.obs; @@ -119,7 +123,6 @@ class TranslationController extends GetxController { Future changeTranslationMode(String mode) async { currentMode.value = mode; currentModeTitle.value = _getModeTitle(mode); - stopAll(); } @@ -137,12 +140,57 @@ class TranslationController extends GetxController { } } + /// 暂停计时器 + void pauseRecordTimer() { + if (_recordTimer != null && !_isTimerPaused) { + _recordTimer?.cancel(); + _recordTimer = null; + _isTimerPaused = true; + Logger.info('录音计时器已暂停'); + } + } + + /// 恢复计时器 + void resumeRecordTimer() { + if (_isTimerPaused) { + _recordTimer = Timer.periodic(Duration(seconds: 1), (_) { + _seconds++; + final minutes = (_seconds ~/ 60).toString().padLeft(2, '0'); + final seconds = (_seconds % 60).toString().padLeft(2, '0'); + recordDurationText.value = '$minutes:$seconds'; + }); + _isTimerPaused = false; + Logger.info('录音计时器已恢复'); + } + } + + /// 重置计时器 + void resetRecordTimer() { + _recordTimer?.cancel(); + _recordTimer = null; + _seconds = 0; + _isTimerPaused = false; + recordDurationText.value = '00:00'; + Logger.info('录音计时器已重置'); + } + Future setActiveSpeaker(int id) async { activeSpeaker.value = id; + await _asrService.startContinuousRecognition(_audioSourceType); // 监听识别结果 if (activeSpeaker.value != 0) { isPlayback = false; - await _asrService.startContinuousRecognition(_audioSourceType); + isRecognizing.value = true; + if (isRecording.value && isRecognizing.value) { + if (lasyIsRecording.value != isRecording.value) { + lasyIsRecording.value = isRecording.value; + await startRecording(); + } else { + await _asrService.resumeRecord(); + // 恢复计时器 + resumeRecordTimer(); + } + } // 开始ASR活跃时长跟踪（面对面模式） _startAsrActiveTracking(); @@ -151,7 +199,11 @@ class TranslationController extends GetxController { } else { // 停止ASR活跃时长跟踪（面对面模式） _stopAsrActiveTracking(); - + if (isRecording.value && isRecognizing.value) { + // 暂停计时器 + pauseRecordTimer(); + await _asrService.pauseRecord(); + } if (fTFTranslationResult != '' && !isPlayback) { isPlayback = true; await _ttsService.setAudioOutputDevice(lastActiveSpeaker.value); @@ -198,6 +250,9 @@ class TranslationController extends GetxController { super.onInit(); //_setupEventListeners(); + // 预先检查权限，避免在用户交互时阻塞 + _preCheckPermissions(); + // 初始化统计服务 _initUsageService(); @@ -228,6 +283,15 @@ class TranslationController extends GetxController { // 参考Agent模块的实现模式 } + /// 预先检查权限 + Future _preCheckPermissions() async { + try { + await requestRecordPermission(); + } catch (e) { + Logger.e("Permission", "预检查权限失败: $e"); + } + } + // 初始化统计服务 void _initUsageService() { try { @@ -264,16 +328,17 @@ class TranslationController extends GetxController { _storage.write(_targetLanguageKey, targetLanguage.value); } - Future _requestRecordPermission() async { + Future requestRecordPermission() async { try { - final bool granted = await PermissionUtil.instance.requestPermission( + hasRecordPermission.value = + await PermissionUtil.instance.requestPermission( permissionType: Permission.microphone, permissionName: 'microphonePermission'.tr, explanationText: 'microphonePermissionExplanation'.tr, permanentDenialText: 'microphonePermissionPermanentDenial'.tr, ); - if (granted) { + if (hasRecordPermission.value) { Logger.d("Permission", "用户授予了录音权限"); return true; } else { @@ -539,11 +604,6 @@ class TranslationController extends GetxController { } Future startRecording() async { - final bool storage = await _requestStoragePermission(); - if (!storage) { - Logger.d("Permission", "未授予存储权限，无法提供录音功能"); - return; - } //isRecording.value = true; _seconds = 0; recordDurationText.value = '00:00'; @@ -559,25 +619,29 @@ class TranslationController extends GetxController { final formattedTime = DateFormat('yyyyMMdd_HHmmss').format(DateTime.now()); await _asrService.enableRecord( "${dir.path}/${currentModeTitle.value.tr}_$formattedTime.wav"); - await _astService.enableRecord( - "${dir.path}/${currentModeTitle.value.tr}_${formattedTime}_mic.wav"); + // await _astService.enableRecord( + // "${dir.path}/${currentModeTitle.value.tr}_${formattedTime}_mic.wav"); + return; } Future stopRecording() async { _recordTimer?.cancel(); _recordTimer = null; isRecording.value = false; + _isTimerPaused = false; recordDurationText.value = '00:00'; - await _asrService.stopRecord(true); - // 停止实际录音逻辑 - // ... + _seconds = 0; + + _asrService.stopRecord(true); + lasyIsRecording.value = false; + Logger.info('录音已停止，计时器已重置'); } // 开始语音识别 Future startRecognition() async { Logger.info('开始语音识别'); // 等待权限请求完成并获取结果 - final bool record = await _requestRecordPermission(); + final bool record = await requestRecordPermission(); // 如果权限被拒绝，直接返回 if (!record) { @@ -634,18 +698,14 @@ class TranslationController extends GetxController { isTtsEnabled.value = true; } - if (currentMode.value == 'faceToFace') { - _asrService.stopContinuousRecognition(); + if (_bluetoothManager.currentDeviceRx.value != null) { + restoreOriginalAudioState(); } else { - if (_bluetoothManager.currentDeviceRx.value != null) { - restoreOriginalAudioState(); - } else { - disableBluetoothAudio(); - } - await _asrService.startContinuousRecognition(_audioSourceType); - // 开始ASR活跃时长计时 - _startAsrActiveTracking(); + disableBluetoothAudio(); } + await _asrService.startContinuousRecognition(_audioSourceType); + // 开始ASR活跃时长计时 + _startAsrActiveTracking(); isRecognizing.value = true; if (isRecording.value && isRecognizing.value) { startRecording(); @@ -691,6 +751,7 @@ class TranslationController extends GetxController { if (!isRecognizing.value) return; try { + await stopRecording(); await _asrService.stopContinuousRecognition(); await _astService.stopContinuousTranslation(); // 停止ASR活跃时长计时 @@ -715,7 +776,7 @@ class TranslationController extends GetxController { saveTranslationHistory(); } - stopRecording(); + currentSessionId = null; } catch (e) { Logger.error('停止语音识别失败: ${e.toString()}'); @@ -993,7 +1054,6 @@ class TranslationController extends GetxController { } // 确保停止ASR活跃时长跟踪 _stopAsrActiveTracking(); - stopRecording(); _translationDebounceTimer?.cancel(); await _ttsService.stop(); } catch (e) { diff --git a/lib/modules/translation/views/translation_view.dart b/lib/modules/translation/views/translation_view.dart index ea8a2f9e1..2998bc6ab 100644 --- a/lib/modules/translation/views/translation_view.dart +++ b/lib/modules/translation/views/translation_view.dart @@ -472,8 +472,8 @@ class TranslationView extends GetView { isDarkMode: isDarkMode, speakerId: 1, activeSpeaker: controller.activeSpeaker.value, - icon: Icons.headset, // 耳机图标 - label: "headsetUser".tr, // 耳机用户 + icon: Icons.headset, + label: "headsetUser".tr, ), SizedBox(width: 40.w), // 手机图标按钮 (说话者2) @@ -481,8 +481,8 @@ class TranslationView extends GetView { isDarkMode: isDarkMode, speakerId: 2, activeSpeaker: controller.activeSpeaker.value, - icon: Icons.phone_android, // 手机图标 - label: "phoneUser".tr, // 手机用户 + icon: Icons.phone_android, + label: "phoneUser".tr, ), ], ) @@ -648,8 +648,13 @@ class TranslationView extends GetView { bool isDisabled = activeSpeaker != 0 && activeSpeaker != speakerId; Timer? _holdTimer; // 增加延迟触发的计时器 return Listener( - onPointerDown: (_) { - // 设置300ms延迟防止误触 + onPointerDown: (_) async { + // 同步检查权限状态，不使用 await + if (!controller.hasRecordPermission.value) { + Get.snackbar('权限提示', '需要麦克风权限才能使用录音功能'); + return; + } + _holdTimer = Timer(const Duration(milliseconds: 200), () { if (activeSpeaker == 0) { controller.setActiveSpeaker(speakerId); diff --git a/local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/AzureAsrHelper.kt b/local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/AzureAsrHelper.kt index 373c2a5d7..048d6d74c 100644 --- a/local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/AzureAsrHelper.kt +++ b/local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/AzureAsrHelper.kt @@ -11,7 +11,7 @@ import android.media.audiofx.AutomaticGainControl import android.util.Log import com.microsoft.cognitiveservices.speech.* import com.microsoft.cognitiveservices.speech.util.EventHandler - +import com.yunqiinnovation.azure_speech.tools.SimpleAudioReceiver import android.media.AudioManager import android.media.AudioAttributes import android.media.AudioFocusRequest @@ -243,14 +243,19 @@ class AzureAsrHelper(private val context: Context) { stopContinuousRecognition() } - - try { - // 启动音频处理 - audioStream!!.startAudioRecord( when(audioSourceType) { - AudioSourceType.MICROPHONE -> SimpleAudioReceiver.AudioSourceType.MICROPHONE - AudioSourceType.EXTERNAL -> SimpleAudioReceiver.AudioSourceType.EXTERNAL - }) + // 启动音频处理（若已启动则跳过） + if (!audioStream!!.isWriting()) { + audioStream!!.startAudioRecord( + when (audioSourceType) { + AudioSourceType.MICROPHONE -> SimpleAudioReceiver.AudioSourceType.MICROPHONE + AudioSourceType.EXTERNAL -> SimpleAudioReceiver.AudioSourceType.EXTERNAL + }, + null + ) + } else { + Log.d(tag, "音频已启动，跳过重复启动") + } // 执行同步识别 val result = recognizer?.recognizeOnceAsync()?.get() @@ -273,13 +278,13 @@ class AzureAsrHelper(private val context: Context) { /** * 开始连续语音识别 - * - * @param callback 连续识别结果回调 - * @return 是否成功开始识别 + * @param audioSourceType 音频源类型 + * @param audioDataCallback 音频数据回调接口 + * @return 是否成功启动 */ fun startContinuousRecognition( - - audioSourceType: AudioSourceType = AudioSourceType.MICROPHONE + audioSourceType: AudioSourceType = AudioSourceType.MICROPHONE, + audioDataCallback: SimpleAudioReceiver.AudioDataCallback? = null ): Boolean { Log.d(tag, "startContinuousRecognition:$isContinuousRecognitionActive ") @@ -295,31 +300,29 @@ class AzureAsrHelper(private val context: Context) { } this.audioSourceType = audioSourceType - try { Log.d(tag, "startContinuousRecognition: ") - // 启动音频处理 - //startAudioProcessing() - // 开始连续识别 + // 启动音频处理（若已启动则跳过） + if (!audioStream!!.isWriting()) { + audioStream!!.startAudioRecord( + when (audioSourceType) { + AudioSourceType.MICROPHONE -> SimpleAudioReceiver.AudioSourceType.MICROPHONE + AudioSourceType.EXTERNAL -> SimpleAudioReceiver.AudioSourceType.EXTERNAL + }, + audioDataCallback + ) + } else { + Log.d(tag, "音频已启动，跳过重复启动") + } - // 启动音频处理 - audioStream!!.startAudioRecord( - when(audioSourceType) { - AudioSourceType.MICROPHONE -> SimpleAudioReceiver.AudioSourceType.MICROPHONE - AudioSourceType.EXTERNAL -> SimpleAudioReceiver.AudioSourceType.EXTERNAL - } - ) recognizer?.startContinuousRecognitionAsync() isContinuousRecognitionActive = true - - return true } catch (e: Exception) { audioStream!!.stopMicrophoneCapture() isContinuousRecognitionActive = false - return false } } @@ -546,7 +549,7 @@ class AzureAsrHelper(private val context: Context) { /** * 开启录音 */ - fun enableRecord(filePath: String) { + fun enableRecord(filePath: String, audioDataCallback: SimpleAudioReceiver.AudioDataCallback? = null) { Log.i(tag, "开启录音:") if (audioStream == null) { // 创建外部音频拉流对象 @@ -555,13 +558,13 @@ class AzureAsrHelper(private val context: Context) { } if (audioStream!!.recordfile == null) { audioStream!!.recordfile = RecordFile() - } + } // 启动音频处理 audioStream!!.startAudioRecord( when(audioSourceType) { AudioSourceType.MICROPHONE -> SimpleAudioReceiver.AudioSourceType.MICROPHONE AudioSourceType.EXTERNAL -> SimpleAudioReceiver.AudioSourceType.EXTERNAL - }) + },audioDataCallback) audioStream!!.recordfile!!.closeFile(true) @@ -618,13 +621,16 @@ class AzureAsrHelper(private val context: Context) { */ fun stopRecord(isSave: Boolean) { Log.i(tag, "关闭录音:") + if (audioStream == null) { + return + } if (audioStream!!.recordfile == null)return audioStream!!.stopMicrophoneCapture() audioStream!!.recordfile!!.closeFile(isSave) } - + /** * 一次性识别回调接口 */ diff --git a/local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/AzureAsrToAsr.kt b/local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/AzureAsrToAsr.kt index d75afa7b6..38c374238 100644 --- a/local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/AzureAsrToAsr.kt +++ b/local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/AzureAsrToAsr.kt @@ -24,6 +24,7 @@ import java.util.* import javax.crypto.Mac import javax.crypto.spec.SecretKeySpec import com.yunqiinnovation.azure_speech.tools.RecordFile +import com.yunqiinnovation.azure_speech.tools.SimpleAudioPlayer /** * 整合的语音翻译服务 * 集成ASR语音识别、翻译服务和TTS语音合成 diff --git a/local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/AzureSpeechPlugin.kt b/local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/AzureSpeechPlugin.kt index 41eb3b938..c936703af 100644 --- a/local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/AzureSpeechPlugin.kt +++ b/local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/AzureSpeechPlugin.kt @@ -5,6 +5,8 @@ import android.os.Handler import android.os.Looper import androidx.annotation.NonNull import com.yunqiinnovation.azure_speech.utils.FileLogger +import com.yunqiinnovation.azure_speech.tools.SimpleAudioReceiver +import com.yunqiinnovation.azure_speech.tools.SimpleAudioPlayer import io.flutter.embedding.engine.plugins.FlutterPlugin import io.flutter.plugin.common.MethodCall import io.flutter.plugin.common.MethodChannel @@ -42,6 +44,12 @@ class AzureSpeechPlugin : BleService.Callback, FlutterPlugin { private var astEventSink: EventChannel.EventSink? = null private lateinit var azureAstHelper: IntegratedSpeechTranslationService + // 音频数据相关 + private lateinit var audioDataChannel: MethodChannel + private lateinit var audioDataEventChannel: EventChannel + private var audioDataEventSink: EventChannel.EventSink? = null + + // 是否已添加TTS事件监听器 private var isTtsListenerAdded = false @@ -96,7 +104,21 @@ class AzureSpeechPlugin : BleService.Callback, FlutterPlugin { } } } + // 音频数据 事件发送方法 + private fun sendAudioDataEvent(event: Map) { + if (audioDataEventSink == null) { + FileLogger.w(tag, "无法发送音频数据事件：事件通道未准备好") + return + } + mainHandler.post { + try { + audioDataEventSink?.success(event) + } catch (e: Exception) { + FileLogger.e(tag, "发送音频数据事件失败: ${e.message}") + } + } + } override fun onAttachedToEngine(@NonNull flutterPluginBinding: FlutterPlugin.FlutterPluginBinding) { context = flutterPluginBinding.applicationContext @@ -149,7 +171,18 @@ class AzureSpeechPlugin : BleService.Callback, FlutterPlugin { astEventSink = null } }) + // 初始化音频数据事件通道 + audioDataEventChannel = + EventChannel(flutterPluginBinding.binaryMessenger, "azure_speech/audio_data_events") + audioDataEventChannel.setStreamHandler(object : EventChannel.StreamHandler { + override fun onListen(arguments: Any?, events: EventChannel.EventSink?) { + audioDataEventSink = events + } + override fun onCancel(arguments: Any?) { + audioDataEventSink = null + } + }) // 初始化Azure语音服务 azureTtsHelper = AzureTtsHelper(context) azureAsrHelper = AzureAsrHelper(context) @@ -473,10 +506,30 @@ class AzureSpeechPlugin : BleService.Callback, FlutterPlugin { "enableRecord" -> { val filePath = call.argument("filePath") ?: "" + // 是否接受音频数据 + val acceptAudioData = call.argument("acceptAudioData") ?: false try { - FileLogger.d(tag, "音频文件名称为: ${filePath}") // - - azureAsrHelper.enableRecord(filePath) + FileLogger.d(tag, "音频文件名称为: ${filePath}") + + // 启用录音，并根据 acceptAudioData 参数决定是否设置音频数据回调 + azureAsrHelper.enableRecord(filePath, if(acceptAudioData) { + // 创建音频数据回调，将音频数据发送到 Flutter 层 + object : SimpleAudioReceiver.AudioDataCallback { + override fun onAudio(audioData: ByteArray) { + // 构建音频数据事件映射 + val audioEvent = mapOf( + "type" to "audioData", + "data" to audioData, + "timestamp" to System.currentTimeMillis(), + "size" to audioData.size + ) + // 发送音频数据事件到 Flutter 层 + sendAudioDataEvent(audioEvent) + } + } + } else { + null + }) result.success(true) } catch (e: Exception) { result.error("ENABLERECORD_ERROR", e.message, null) diff --git a/local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/AzureTtsHelper.kt b/local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/AzureTtsHelper.kt index 709b99e66..0c0600aba 100644 --- a/local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/AzureTtsHelper.kt +++ b/local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/AzureTtsHelper.kt @@ -14,7 +14,7 @@ import com.deep_voice.speech.tts.TtsEventType import kotlinx.coroutines.* import java.io.ByteArrayInputStream import java.io.InputStream - +import com.yunqiinnovation.azure_speech.tools.SimpleAudioPlayer /** * Azure TTS Helper * diff --git a/local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/tools/SimpleAudioPlayer.kt b/local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/tools/SimpleAudioPlayer.kt index 61870debb..65727a25d 100644 --- a/local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/tools/SimpleAudioPlayer.kt +++ b/local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/tools/SimpleAudioPlayer.kt @@ -1,4 +1,4 @@ -package com.yunqiinnovation.azure_speech +package com.yunqiinnovation.azure_speech.tools import android.content.Context import android.media.AudioAttributes diff --git a/local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/tools/SimpleAudioReceiver.kt b/local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/tools/SimpleAudioReceiver.kt index 75f76d53a..c9d50f07e 100644 --- a/local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/tools/SimpleAudioReceiver.kt +++ b/local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/tools/SimpleAudioReceiver.kt @@ -1,4 +1,4 @@ -package com.yunqiinnovation.azure_speech +package com.yunqiinnovation.azure_speech.tools import android.content.Context import android.media.AudioFormat @@ -38,6 +38,7 @@ class SimpleAudioReceiver(private val context: Context) { } private val bufferSize = 4096 // 可根据需要调整 + private var audioDataCallback: AudioDataCallback? = null var audioRecord: AudioRecord? = null var pushAudioStream: PushAudioInputStream? = null // 音频源配置 @@ -98,7 +99,7 @@ class SimpleAudioReceiver(private val context: Context) { * 启动写入线程 */ private fun startWriteThread() { - + writeThread = Thread { try { while (isRunning.get()) { @@ -139,6 +140,9 @@ class SimpleAudioReceiver(private val context: Context) { try { Log.d(TAG, "写入数据大小: ${finalData.size}") + if(audioDataCallback!=null){ + audioDataCallback?.onAudio(finalData) + } pushAudioStream?.write(finalData) @@ -175,12 +179,15 @@ class SimpleAudioReceiver(private val context: Context) { /** * 开始音频输入 + * @param audioSourceType 音频源类型 + * @param callback 音频数据回调，可为空 */ - fun startAudioRecord(audioSourceType: AudioSourceType) { + fun startAudioRecord(audioSourceType: AudioSourceType, callback: AudioDataCallback?) { this.audioSourceType = audioSourceType + this.audioDataCallback = callback writeQueue.clear() isWriting.set(true) - + when (audioSourceType) { AudioSourceType.MICROPHONE -> runMicrophoneCapture() AudioSourceType.EXTERNAL -> runExternalCapture() @@ -400,4 +407,20 @@ class SimpleAudioReceiver(private val context: Context) { Log.e("AudioMode", "恢复音频模式失败: ${e.message}") } } + + fun isWriting(): Boolean { + return isWriting.get() + } + + /** + * 连续识别回调接口 + */ + interface AudioDataCallback { + /** + * 音频数据回调 + * @param data 音频数据字节数组 + */ + fun onAudio(data: ByteArray) + } } +