From dc9c3a42761499d96f066d8479150c22e2a0bf5f Mon Sep 17 00:00:00 2001 From: wolfplus2048 <30993207+wolfplus2048@users.noreply.github.com> Date: Sat, 21 Jun 2025 22:02:59 +0100 Subject: [PATCH] add --- .../azure_speech/AzureSpeechPlugin.kt | 8 ++ .../azure_speech/AzureTtsHelper.kt | 44 +++++++- .../bytedance_speech/BytedanceAudioPlayer.kt | 100 ++++++++++++++++-- .../bytedance_speech/BytedanceSpeechPlugin.kt | 6 ++ .../bytedance_speech/BytedanceTTS.kt | 24 ++++- .../com/deep_voice/speech/ITtsService.kt | 7 ++ .../kotlin/com/deep_voice/speech/TtsEvents.kt | 16 +++ 7 files changed, 193 insertions(+), 12 deletions(-) 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 e19579cc3..54e960092 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 @@ -158,6 +158,14 @@ class AzureSpeechPlugin : BleService.Callback, FlutterPlugin { ) } + TtsEventType.PLAYBACK_STARTED -> mapOf( + "type" to "playback_started" + ) + + TtsEventType.PLAYBACK_COMPLETED -> mapOf( + "type" to "playback_completed" + ) + TtsEventType.ERROR -> { val params = event.params mapOf( 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 9b3d8ccfe..0465e5e4c 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 @@ -1,14 +1,16 @@ package com.yunqiinnovation.azure_speech import android.content.Context +import android.media.AudioManager import com.microsoft.cognitiveservices.speech.* import com.microsoft.cognitiveservices.speech.audio.* import com.yunqiinnovation.azure_speech.utils.FileLogger +import com.deep_voice.speech.tts.AudioDataListener +import com.deep_voice.speech.tts.AudioOutputDevice import com.deep_voice.speech.tts.ITtsService import com.deep_voice.speech.tts.TtsEvent import com.deep_voice.speech.tts.TtsEventListener import com.deep_voice.speech.tts.TtsEventType -import com.deep_voice.speech.tts.AudioDataListener import kotlinx.coroutines.* import java.io.ByteArrayInputStream import java.io.InputStream @@ -203,6 +205,42 @@ class AzureTtsHelper(private val context: Context) : ITtsService { } } + /** + * 设置音频输出设备 + * + * @param device 音频输出设备类型 + */ + override fun setAudioOutputDevice(device: AudioOutputDevice) { + // Azure TTS使用系统默认的音频路由 + // 音频输出设备的控制需要通过Android的AudioManager实现 + val audioManager = context.getSystemService(Context.AUDIO_SERVICE) as? AudioManager + audioManager?.let { manager -> + when (device) { + AudioOutputDevice.DEFAULT -> { + // 默认模式：系统自动选择 + manager.mode = AudioManager.MODE_NORMAL + manager.isSpeakerphoneOn = false + } + AudioOutputDevice.SPEAKER -> { + // 强制使用扬声器 + manager.mode = AudioManager.MODE_NORMAL + manager.isSpeakerphoneOn = true + } + AudioOutputDevice.HEADPHONES -> { + // 强制使用耳机（如果已连接） + manager.mode = AudioManager.MODE_NORMAL + manager.isSpeakerphoneOn = false + // 注意：Android不能强制路由到耳机，只能在耳机已连接时使用 + } + AudioOutputDevice.EARPIECE -> { + // 强制使用听筒 + manager.mode = AudioManager.MODE_IN_COMMUNICATION + manager.isSpeakerphoneOn = false + } + } + } + } + /** * 触发事件通知 */ @@ -228,6 +266,8 @@ class AzureTtsHelper(private val context: Context) : ITtsService { SynthesisStarted?.addEventListener { _, eventArgs -> FileLogger.d(TAG, "语音合成开始: resultId=${eventArgs.result.resultId}") notifyEvent(TtsEventType.SYNTHESIS_STARTED) + // Azure TTS 在使用默认音频输出时会立即开始播放 + notifyEvent(TtsEventType.PLAYBACK_STARTED) } // 合成中事件（接收音频数据） @@ -253,6 +293,8 @@ class AzureTtsHelper(private val context: Context) : ITtsService { FileLogger.d(TAG, "语音合成完成: resultId=${eventArgs.result.resultId}, 音频长度=${eventArgs.result.audioLength} 字节") isSpeaking = false notifyEvent(TtsEventType.SYNTHESIS_COMPLETED) + // Azure TTS 合成完成即播放完成 + notifyEvent(TtsEventType.PLAYBACK_COMPLETED) } // 合成取消事件 diff --git a/local_plugins/bytedance_speech/android/src/main/kotlin/com/deep_voice/bytedance_speech/BytedanceAudioPlayer.kt b/local_plugins/bytedance_speech/android/src/main/kotlin/com/deep_voice/bytedance_speech/BytedanceAudioPlayer.kt index 84e73b1be..a3f2c972b 100644 --- a/local_plugins/bytedance_speech/android/src/main/kotlin/com/deep_voice/bytedance_speech/BytedanceAudioPlayer.kt +++ b/local_plugins/bytedance_speech/android/src/main/kotlin/com/deep_voice/bytedance_speech/BytedanceAudioPlayer.kt @@ -1,5 +1,6 @@ package com.deep_voice.bytedance_speech +import android.content.Context import android.media.AudioAttributes import android.media.AudioFormat import android.media.AudioManager @@ -7,12 +8,13 @@ import android.media.AudioTrack import android.os.Build import android.util.Log import com.deep_voice.speech.tts.AudioDataListener +import com.deep_voice.speech.tts.AudioOutputDevice /** * 简化的音频流播放器 * 基于位置标记触发播放完成回调 */ -class BytedanceAudioPlayer : AudioDataListener { +class BytedanceAudioPlayer(private val context: Context) : AudioDataListener { companion object { private const val TAG = "BytedanceAudioPlayer" private const val SAMPLE_RATE = 24000 @@ -24,6 +26,8 @@ class BytedanceAudioPlayer : AudioDataListener { private var isFirstData = true // 是否是第一次接收数据 private var totalBytesWritten = 0 // 总共写入的字节数 private var sessionActive = true // 会话是否活跃 + private var audioOutputDevice = AudioOutputDevice.DEFAULT // 音频输出设备 + private var audioManager: AudioManager? = null // 回调 private var onPlayStarted: (() -> Unit)? = null @@ -40,6 +44,12 @@ class BytedanceAudioPlayer : AudioDataListener { // 不使用周期性通知 } } + + init { + Log.d(TAG, "BytedanceAudioPlayer 初始化") + audioManager = context.getSystemService(Context.AUDIO_SERVICE) as? AudioManager + initAudioTrack() + } /** @@ -198,6 +208,23 @@ class BytedanceAudioPlayer : AudioDataListener { return audioTrack?.playState == AudioTrack.PLAYSTATE_PLAYING } + /** + * 设置音频输出设备 + */ + fun setAudioOutputDevice(device: AudioOutputDevice) { + if (audioOutputDevice != device) { + audioOutputDevice = device + // 如果AudioTrack已初始化，需要重新创建以应用新的输出设备设置 + if (audioTrack != null) { + val wasPlaying = isPlaying() + initAudioTrack() + if (wasPlaying) { + audioTrack?.play() + } + } + } + } + /** * 初始化 AudioTrack */ @@ -209,15 +236,36 @@ class BytedanceAudioPlayer : AudioDataListener { val minBufferSize = AudioTrack.getMinBufferSize(SAMPLE_RATE, CHANNEL_CONFIG, AUDIO_FORMAT) val bufferSize = minBufferSize * 2 - // 创建 AudioTrack - audioTrack = if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.M) { - AudioTrack.Builder() - .setAudioAttributes( + // 根据输出设备配置AudioAttributes + val audioAttributes = when (audioOutputDevice) { + AudioOutputDevice.EARPIECE -> { + // 听筒模式 + if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.M) { + AudioAttributes.Builder() + .setUsage(AudioAttributes.USAGE_VOICE_COMMUNICATION) + .setContentType(AudioAttributes.CONTENT_TYPE_SPEECH) + .build() + } else { + null + } + } + else -> { + // 默认、耳机、扬声器模式 + if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.M) { AudioAttributes.Builder() .setUsage(AudioAttributes.USAGE_MEDIA) .setContentType(AudioAttributes.CONTENT_TYPE_SPEECH) .build() - ) + } else { + null + } + } + } + + // 创建 AudioTrack + audioTrack = if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.M && audioAttributes != null) { + AudioTrack.Builder() + .setAudioAttributes(audioAttributes) .setAudioFormat( AudioFormat.Builder() .setEncoding(AUDIO_FORMAT) @@ -230,8 +278,12 @@ class BytedanceAudioPlayer : AudioDataListener { .build() } else { @Suppress("DEPRECATION") + val streamType = when (audioOutputDevice) { + AudioOutputDevice.EARPIECE -> AudioManager.STREAM_VOICE_CALL + else -> AudioManager.STREAM_MUSIC + } AudioTrack( - AudioManager.STREAM_MUSIC, + streamType, SAMPLE_RATE, CHANNEL_CONFIG, AUDIO_FORMAT, @@ -243,9 +295,43 @@ class BytedanceAudioPlayer : AudioDataListener { // 设置播放位置监听器 audioTrack?.setPlaybackPositionUpdateListener(playbackListener) + // 配置音频路由 + configureAudioRouting() + // 开始播放 audioTrack?.play() Log.d(TAG, "AudioTrack 初始化成功") } + + /** + * 配置音频路由 + */ + private fun configureAudioRouting() { + audioManager?.let { manager -> + when (audioOutputDevice) { + AudioOutputDevice.DEFAULT -> { + // 默认模式：系统自动选择 + manager.mode = AudioManager.MODE_NORMAL + manager.isSpeakerphoneOn = false + } + AudioOutputDevice.SPEAKER -> { + // 强制使用扬声器 + manager.mode = AudioManager.MODE_NORMAL + manager.isSpeakerphoneOn = true + } + AudioOutputDevice.HEADPHONES -> { + // 强制使用耳机（如果已连接） + manager.mode = AudioManager.MODE_NORMAL + manager.isSpeakerphoneOn = false + // 注意：Android不能强制路由到耳机，只能在耳机已连接时使用 + } + AudioOutputDevice.EARPIECE -> { + // 强制使用听筒 + manager.mode = AudioManager.MODE_IN_COMMUNICATION + manager.isSpeakerphoneOn = false + } + } + } + } } \ No newline at end of file diff --git a/local_plugins/bytedance_speech/android/src/main/kotlin/com/deep_voice/bytedance_speech/BytedanceSpeechPlugin.kt b/local_plugins/bytedance_speech/android/src/main/kotlin/com/deep_voice/bytedance_speech/BytedanceSpeechPlugin.kt index d86faa71f..a6003e050 100644 --- a/local_plugins/bytedance_speech/android/src/main/kotlin/com/deep_voice/bytedance_speech/BytedanceSpeechPlugin.kt +++ b/local_plugins/bytedance_speech/android/src/main/kotlin/com/deep_voice/bytedance_speech/BytedanceSpeechPlugin.kt @@ -255,6 +255,12 @@ class BytedanceSpeechPlugin : FlutterPlugin { TtsEventType.SYNTHESIS_CANCELED -> { sendTTSEvent("canceled", null) } + TtsEventType.PLAYBACK_STARTED -> { + sendTTSEvent("playback_started", null) + } + TtsEventType.PLAYBACK_COMPLETED -> { + sendTTSEvent("playback_completed", null) + } TtsEventType.ERROR -> { val params = HashMap() params["code"] = event.params["errorCode"] ?: "UNKNOWN_ERROR" diff --git a/local_plugins/bytedance_speech/android/src/main/kotlin/com/deep_voice/bytedance_speech/BytedanceTTS.kt b/local_plugins/bytedance_speech/android/src/main/kotlin/com/deep_voice/bytedance_speech/BytedanceTTS.kt index 49d83efcc..b3cd7e75f 100644 --- a/local_plugins/bytedance_speech/android/src/main/kotlin/com/deep_voice/bytedance_speech/BytedanceTTS.kt +++ b/local_plugins/bytedance_speech/android/src/main/kotlin/com/deep_voice/bytedance_speech/BytedanceTTS.kt @@ -3,6 +3,7 @@ package com.deep_voice.bytedance_speech import android.content.Context import android.util.Log import com.deep_voice.speech.tts.AudioDataListener +import com.deep_voice.speech.tts.AudioOutputDevice import com.deep_voice.speech.tts.ITtsService import com.deep_voice.speech.tts.TtsEvent import com.deep_voice.speech.tts.TtsEventListener @@ -98,7 +99,7 @@ class BytedanceTTS(private val context: Context) : ITtsService, CoroutineScope { private val audioDataListeners = mutableListOf() // 内部音频播放器 - private val audioPlayer = BytedanceAudioPlayer() + private val audioPlayer = BytedanceAudioPlayer(context) // 是否使用内部播放器 private var useInternalPlayer = true @@ -159,14 +160,14 @@ class BytedanceTTS(private val context: Context) : ITtsService, CoroutineScope { audioPlayer.setOnPlayStarted { Log.d(TAG, "播放开始") updateStatus(STATUS_SPEAKING) - notifyEvent(TtsEventType.SYNTHESIS_STARTED) + notifyEvent(TtsEventType.PLAYBACK_STARTED) } // 设置播放完成回调 audioPlayer.setOnPlayCompleted { Log.d(TAG, "播放结束") updateStatus(STATUS_READY) - notifyEvent(TtsEventType.SYNTHESIS_COMPLETED) + notifyEvent(TtsEventType.PLAYBACK_COMPLETED) } // 设置错误回调 @@ -421,6 +422,15 @@ class BytedanceTTS(private val context: Context) : ITtsService, CoroutineScope { updateInternalPlayerUsage() } + /** + * 设置音频输出设备 + */ + override fun setAudioOutputDevice(device: AudioOutputDevice) { + if (useInternalPlayer) { + audioPlayer.setAudioOutputDevice(device) + } + } + /** * 更新内部播放器的使用状态 */ @@ -650,7 +660,9 @@ class BytedanceTTS(private val context: Context) : ITtsService, CoroutineScope { EVENT_SESSION_STARTED -> { connectionAttempts = 0 - // 会话开始时，如果使用内部播放器则启动播放器会话 + // 会话开始时触发合成开始事件 + notifyEvent(TtsEventType.SYNTHESIS_STARTED) + // 如果使用内部播放器则启动播放器会话 if (useInternalPlayer) { audioPlayer.startSession() } @@ -672,6 +684,10 @@ class BytedanceTTS(private val context: Context) : ITtsService, CoroutineScope { EVENT_SESSION_FINISHED -> { isSessionStarted = false Log.i(TAG, "EVENT_SESSION_FINISHED, ${sessionId}") + + // 触发合成完成事件 + notifyEvent(TtsEventType.SYNTHESIS_COMPLETED) + // 如果使用内部播放器则结束播放器会话 if (useInternalPlayer) { audioPlayer.endSession() diff --git a/local_plugins/speech/android/src/main/kotlin/com/deep_voice/speech/ITtsService.kt b/local_plugins/speech/android/src/main/kotlin/com/deep_voice/speech/ITtsService.kt index d43604582..00db2f2b3 100644 --- a/local_plugins/speech/android/src/main/kotlin/com/deep_voice/speech/ITtsService.kt +++ b/local_plugins/speech/android/src/main/kotlin/com/deep_voice/speech/ITtsService.kt @@ -97,4 +97,11 @@ interface ITtsService { * false: 仅通过音频数据监听器输出数据，不播放 */ fun setUseInternalPlayer(useInternalPlayer: Boolean) + + /** + * 设置音频输出设备 + * + * @param device 音频输出设备类型 + */ + fun setAudioOutputDevice(device: AudioOutputDevice) } \ No newline at end of file diff --git a/local_plugins/speech/android/src/main/kotlin/com/deep_voice/speech/TtsEvents.kt b/local_plugins/speech/android/src/main/kotlin/com/deep_voice/speech/TtsEvents.kt index cb69a849d..6758c6f05 100644 --- a/local_plugins/speech/android/src/main/kotlin/com/deep_voice/speech/TtsEvents.kt +++ b/local_plugins/speech/android/src/main/kotlin/com/deep_voice/speech/TtsEvents.kt @@ -13,6 +13,12 @@ enum class TtsEventType { /** 合成取消 */ SYNTHESIS_CANCELED, + /** 播放开始 */ + PLAYBACK_STARTED, + + /** 播放结束 */ + PLAYBACK_COMPLETED, + /** 发生错误 */ ERROR } @@ -52,4 +58,14 @@ interface AudioDataListener { * @param data 音频数据字节数组 */ fun onAudioData(data: ByteArray) +} + +/** + * 音频输出设备类型 + */ +enum class AudioOutputDevice { + DEFAULT, // 默认（如果有耳机选耳机，否则使用系统扬声器） + HEADPHONES, // 强制使用耳机 + SPEAKER, // 强制使用扬声器 + EARPIECE // 强制使用听筒 } \ No newline at end of file