修改声音复刻

lib/data/services/speech_impl/azure_tts_service.dart

@@ -182,6 +182,14 @@ class AzureTtsService extends GetxService implements TtsService {
     }
   }
 
+  @override
+  Future<bool> setVoiceFlocking(String speakerProfileId) async {
+    final result = await _channel.invokeMethod('setVoiceFlocking', {
+      'speakerProfileId': speakerProfileId,
+    });
+    return result;
+  }
+
   @override
   Future<bool> startspeak(String sessionid) async {
     if (!_isInitialized) await initialize();

lib/data/services/speech_impl/voice_clone_tts_service.dart

@@ -94,6 +94,11 @@ class VoiceCloneTtsService extends GetxService implements TtsService {
     }
   }
 
+  @override
+  Future<bool> setVoiceFlocking(String speakerProfileId) async {
+    return true;
+  }
+
   // 语音合成
   Future<void> _speechSynthesis(String text) async {
     await _tryConnecting();

lib/data/services/speech_impl/volcano_tts_service.dart

@@ -185,6 +185,11 @@ class VolcanoTtsService extends GetxService implements TtsService {
     }
   }
 
+  @override
+  Future<bool> setVoiceFlocking(String speakerProfileId) async {
+    return true;
+  }
+
   /// 设置是否启用音频数据回调
   Future<bool> setEnableAudioCallback(bool enable) async {
     if (!await _ensureInitialized()) return false;

lib/data/services/tts_service.dart

@@ -19,6 +19,9 @@ abstract class TtsService {
   /// 设置语音
   Future<bool> setVoice(String voiceName);
 
+  /// 设置语音复刻
+  Future<bool> setVoiceFlocking(String speakerProfileId);
+
   ///开始新的播放会话
   Future<bool> startspeak(String sessionid);
 

lib/modules/translation/middleware/controllers/translation_middleware_controller.dart

@@ -75,16 +75,16 @@ class TranslationMiddlewareController extends GetxController {
       iconColor: Colors.indigo[500]!, // 主靛
       targetRoute: Routes.textTranslation,
     ),
-    // TranslationOption(
+    TranslationOption(
-    //   translationId: 'voice_replication',
+      translationId: 'voice_replication',
-    //   title: '声音复刻',
+      title: '声音复刻',
-    //   icon: Icons.volume_up,
+      icon: Icons.volume_up,
-    //   startColor: Colors.pink[400]!, // 粉色
+      startColor: Colors.pink[400]!, // 粉色
-    //   endColor: Colors.pink[600]!, // 深粉
+      endColor: Colors.pink[600]!, // 深粉
-    //   iconBgColor: Colors.white, // 白色
+      iconBgColor: Colors.white, // 白色
-    //   iconColor: Colors.pink[500]!, // 主粉
+      iconColor: Colors.pink[500]!, // 主粉
-    //   targetRoute: Routes.voiceReplication,
+      targetRoute: Routes.voiceReplication,
-    // ),
+    ),
   ];
 
   /// 导航到指定的翻译功能

lib/modules/voice_replication/views/recorder_bottom_sheet.dart

@@ -118,11 +118,53 @@ class _RecorderBottomSheetState extends State<RecorderBottomSheet> {
 
     final streamed = await request.send();
     final response = await http.Response.fromStream(streamed);
-    if (streamed.statusCode < 200 || streamed.statusCode >= 300) {
+    if (streamed.statusCode >= 200 && streamed.statusCode < 300) {
+      final bodyText = utf8.decode(response.bodyBytes, allowMalformed: true);
+      if (bodyText.isNotEmpty) {
+        final decoded = jsonDecode(bodyText);
+
+        String status = decoded['status'] ?? 'NotStarted';
+        final operationLocation = response.headers['operation-location'];
+        int retryCount = 0;
+
+        while (status != 'Succeeded' && status != 'Failed' && retryCount < 30) {
+          await Future.delayed(const Duration(seconds: 1));
+          if (operationLocation != null && operationLocation.isNotEmpty) {
+            status = await _operations(operationLocation);
+          } else {
+            final consentUrl =
+                'https://$_resourceRegion.api.cognitive.microsoft.com/customvoice/consents/$_jessicaConsentId?api-version=2024-02-01-preview';
+            status = await _operations(consentUrl);
+          }
+          retryCount++;
+        }
+
+        if (status != 'Succeeded') {
+          throw Exception('Consent validation failed or timed out: $status');
+        }
+      }
+    } else {
       throw Exception('Failed to create consent: ${response.statusCode}');
     }
   }
 
+  Future<String> _operations(String operationUrl) async {
+    String status = 'NotStarted';
+
+    final url = Uri.parse(operationUrl);
+    final response = await http.get(url, headers: {
+      'Ocp-Apim-Subscription-Key': _resourceKey,
+    });
+    if (response.statusCode == 200) {
+      final bodyText = utf8.decode(response.bodyBytes, allowMalformed: true);
+      if (bodyText.isNotEmpty) {
+        final decoded = jsonDecode(bodyText);
+        status = decoded['status'] ?? status;
+      }
+    }
+    return status;
+  }
+
   Future<void> _personalvoices() async {
     final url = Uri.parse(
         'https://$_resourceRegion.api.cognitive.microsoft.com/customvoice/personalvoices/$_jessicaPersonalVoiceId?api-version=2024-02-01-preview');
@@ -133,8 +175,6 @@ class _RecorderBottomSheetState extends State<RecorderBottomSheet> {
     request.fields['consentId'] = _jessicaConsentId;
 
     request.files.add(await http.MultipartFile.fromPath('audiodata', _path));
-    request.files.add(await http.MultipartFile.fromPath('audiodata',
-        '/data/user/0/com.saitong.voitrans/app_flutter/VoiceReplication/recording_1764763812656.wav'));
 
     final streamed = await request.send();
     final response = await http.Response.fromStream(streamed);
@@ -142,17 +182,11 @@ class _RecorderBottomSheetState extends State<RecorderBottomSheet> {
       final bodyText = utf8.decode(response.bodyBytes, allowMalformed: true);
       if (bodyText.isNotEmpty) {
         final decoded = jsonDecode(bodyText);
-        print('------2-------$decoded');
         _controller.speakerProfileId.value = decoded['speakerProfileId'];
         _storage.write('open_voice_replication', true);
         _storage.write('speakerProfileId', decoded['speakerProfileId']);
       }
     } else {
-      final bodyText = utf8.decode(response.bodyBytes, allowMalformed: true);
-      if (bodyText.isNotEmpty) {
-        final decoded = jsonDecode(bodyText);
-        print('------2-------$decoded');
-      }
       throw Exception(
           'Failed to create personal voice: ${response.statusCode}');
     }

local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/AzureSpeechPlugin.kt

@@ -699,6 +699,11 @@ class AzureSpeechPlugin : BleService.Callback, FlutterPlugin {
                     result.success(azureTtsHelper.setVoice(voiceName))
                 }
 
+                "setVoiceFlocking" -> {
+                    val speakerProfileId = call.argument<String>("speakerProfileId") ?: ""
+                    result.success(azureTtsHelper.setVoiceFlocking(speakerProfileId))
+                }
+
                 "speakText" -> {
                     val text = call.argument<String>("text") ?: return result.error(
                         "INVALID_ARGUMENTS",

local_plugins/azure_speech/android/src/main/kotlin/com/yunqiinnovation/azure_speech/AzureTtsHelper.kt

@@ -51,6 +51,9 @@ class AzureTtsHelper(private val context: Context) : ITtsService {
     private val streamBuffer = StringBuilder()
     private var lastSpeakTime = 0L
 
+    private var enableVoiceFlocking = false
+    private var speakerProfileId: String = ""
+
     /**
      * 初始化TTS引擎
      *
@@ -354,6 +357,19 @@ class AzureTtsHelper(private val context: Context) : ITtsService {
         }
     }
 
+    /**
+     * 开启声音复刻
+     */
+    fun setVoiceFlocking(id: String): Boolean {
+        if(id.isNotEmpty()) {
+            enableVoiceFlocking = true
+        } else {
+            enableVoiceFlocking = false
+        }
+        speakerProfileId = id
+        return true
+    }
+
     /**
      * 重新创建合成器
      */
@@ -502,16 +518,27 @@ class AzureTtsHelper(private val context: Context) : ITtsService {
             .replace(Regex("[😀-🟿]+"), "") // 简化表情符号移除
             .trim()
 
-        // 优化：简化SSML结构，减少嵌套层级
+        if(enableVoiceFlocking) {
-        return """
+            return """
-        <speak version="1.0" xmlns="http://www.w3.org/2001/10/synthesis" xml:lang="zh-CN">
+            <speak version="1.0" xmlns="http://www.w3.org/2001/10/synthesis" xmlns:mstts="http://www.w3.org/2001/mstts" xml:lang="zh-CN">
-          <voice name="$currentVoice">
+                <voice name="DragonLatestNeural">
-            <prosody rate="$currentRate" pitch="$currentPitch" volume="$currentVolume">
+                    <mstts:ttsembedding speakerProfileId="$speakerProfileId">
-              $processedText
+                        $processedText
-            </prosody>
+                    </mstts:ttsembedding>
-          </voice>
+                </voice>
-        </speak>
+            </speak>
-        """.trimIndent()
+            """.trimIndent()
+        } else {
+            return """
+            <speak version="1.0" xmlns="http://www.w3.org/2001/10/synthesis" xml:lang="zh-CN">
+                <voice name="$currentVoice">
+                    <prosody rate="$currentRate" pitch="$currentPitch" volume="$currentVolume">
+                        $processedText
+                    </prosody>
+                </voice>
+            </speak>
+            """.trimIndent()
+        }
     }
 
     /**