Merge remote-tracking branch 'upstream/py_test_Memory_powermem' into add-powermem

# Conflicts:
#	main/xiaozhi-server/core/providers/memory/powermem/powermem.py

main/manager-api/src/main/java/xiaozhi/modules/model/dto/VoiceDTO.java

@@ -22,12 +22,24 @@ public class VoiceDTO implements Serializable {
 
     @Schema(description = "音频播放地址")
     private String voiceDemo;
+    
+    @Schema(description = "是否为克隆音色")
+    private Boolean isClone;
 
     // 添加双参数构造函数，保持向后兼容
     public VoiceDTO(String id, String name) {
         this.id = id;
         this.name = name;
         this.voiceDemo = null;
+        this.isClone = false; // 默认不是克隆音色
+    }
+    
+    // 添加三参数构造函数，用于普通音色
+    public VoiceDTO(String id, String name, String voiceDemo) {
+        this.id = id;
+        this.name = name;
+        this.voiceDemo = voiceDemo;
+        this.isClone = false;
     }
 
 }

main/manager-api/src/main/java/xiaozhi/modules/timbre/service/impl/TimbreServiceImpl.java

@@ -136,6 +136,7 @@ public class TimbreServiceImpl extends BaseServiceImpl<TimbreDao, TimbreEntity>
                 .map(entity -> {
                     VoiceDTO dto = new VoiceDTO(entity.getId(), entity.getName());
                     dto.setVoiceDemo(entity.getVoiceDemo());
+                    dto.setIsClone(false); // 设置为普通音色
                     return dto;
                 })
                 .collect(Collectors.toList());
@@ -152,6 +153,7 @@ public class TimbreServiceImpl extends BaseServiceImpl<TimbreDao, TimbreEntity>
                 voiceDTO.setName(MessageUtils.getMessage(ErrorCode.VOICE_CLONE_PREFIX) + entity.getName());
                 // 保留从数据库查询到的voiceDemo字段
                 voiceDTO.setVoiceDemo(entity.getVoiceDemo());
+                voiceDTO.setIsClone(true); // 设置为克隆音色
                 redisUtils.set(RedisKeys.getTimbreNameById(voiceDTO.getId()), voiceDTO.getName(),
                         RedisUtils.NOT_EXPIRE);
                 voiceDTOs.add(0, voiceDTO);
@@ -214,6 +216,7 @@ public class TimbreServiceImpl extends BaseServiceImpl<TimbreDao, TimbreEntity>
         TimbreEntity entity = list.get(0);
         VoiceDTO dto = new VoiceDTO(entity.getId(), entity.getName());
         dto.setVoiceDemo(entity.getVoiceDemo());
+        dto.setIsClone(false); // 设置为普通音色
         return dto;
     }
 }

main/manager-web/src/i18n/de.js

@@ -779,6 +779,16 @@ export default {
   'roleConfig.fetchConfigFailed': 'Konfiguration konnte nicht abgerufen werden',
   'roleConfig.fetchModelsFailed': 'Modellliste konnte nicht abgerufen werden',
   'roleConfig.fetchPluginsFailed': 'Plugin-Liste konnte nicht abgerufen werden',
+  'roleConfig.cannotResumeAudio': 'Audio-Wiedergabe kann nicht fortgesetzt werden',
+  'roleConfig.selectVoiceFirst': 'Bitte wählen Sie zuerst eine Stimme aus',
+  'roleConfig.audioLoadTimeout': 'Audio-Laden dauert zu lange, bitte versuchen Sie es später erneut',
+  'roleConfig.cloneAudioPlayFailed': 'Klon-Audio-Wiedergabe fehlgeschlagen',
+  'roleConfig.cannotPlayCloneAudio': 'Klon-Audio kann nicht abgespielt werden',
+  'roleConfig.getCloneAudioFailed': 'Klon-Audio konnte nicht abgerufen werden',
+  'roleConfig.noPreviewAudio': 'Diese Stimme hat keine Vorschau-Audio-Datei',
+  'roleConfig.audioPlayFailed': 'Audio-Wiedergabe fehlgeschlagen',
+  'roleConfig.cannotPlayAudio': 'Audio kann nicht abgespielt werden',
+  'roleConfig.audioPlayError': 'Fehler bei der Audio-Wiedergabe',
 
   // Function management dialog text
   'functionDialog.title': 'Funktionsverwaltung',
@@ -919,6 +929,10 @@ export default {
   'ttsModel.selectVoiceToDelete': 'Bitte wählen Sie Stimmen zum Löschen aus',
   'ttsModel.warning': 'Warnung',
   'ttsModel.confirmDeleteVoice': 'Sind Sie sicher, dass Sie {count} Stimmen löschen möchten?',
+  'ttsModel.voiceCodeNameLanguageRequired': 'Stimmcode, Stimmaname und Sprache können nicht leer sein',
+  'ttsModel.deleteCancelled': 'Löschvorgang abgebrochen',
+  'ttsModel.updateSuccess': 'Aktualisierung erfolgreich',
+  'ttsModel.saveSuccess': 'Speicherung erfolgreich',
 
   // OTA Management Page Text
   'otaManagement.firmwareManagement': 'Firmware-Verwaltung',

main/manager-web/src/i18n/en.js

@@ -779,6 +779,16 @@ export default {
   'roleConfig.fetchConfigFailed': 'Failed to fetch configuration',
   'roleConfig.fetchModelsFailed': 'Failed to fetch model list',
   'roleConfig.fetchPluginsFailed': 'Failed to fetch plugin list',
+  'roleConfig.cannotResumeAudio': 'Cannot resume audio playback',
+  'roleConfig.selectVoiceFirst': 'Please select a voice first',
+  'roleConfig.audioLoadTimeout': 'Audio loading takes too long, please try again later',
+  'roleConfig.cloneAudioPlayFailed': 'Clone audio playback failed',
+  'roleConfig.cannotPlayCloneAudio': 'Cannot play clone audio',
+  'roleConfig.getCloneAudioFailed': 'Failed to get clone audio',
+  'roleConfig.noPreviewAudio': 'This voice has no preview audio available',
+  'roleConfig.audioPlayFailed': 'Audio playback failed',
+  'roleConfig.cannotPlayAudio': 'Cannot play audio',
+  'roleConfig.audioPlayError': 'Error occurred during audio playback',
 
   // Function management dialog text
   'functionDialog.title': 'Function Management',
@@ -919,6 +929,10 @@ export default {
   'ttsModel.selectVoiceToDelete': 'Please select voices to delete',
   'ttsModel.warning': 'Warning',
   'ttsModel.confirmDeleteVoice': 'Are you sure to delete {count} voices?',
+  'ttsModel.voiceCodeNameLanguageRequired': 'Voice code, voice name and language type cannot be empty',
+  'ttsModel.deleteCancelled': 'Deletion cancelled',
+  'ttsModel.updateSuccess': 'Update successful',
+  'ttsModel.saveSuccess': 'Save successful',
 
   // OTA Management Page Text
   'otaManagement.firmwareManagement': 'Firmware Management',

main/manager-web/src/i18n/vi.js

@@ -779,6 +779,16 @@ export default {
   'roleConfig.fetchConfigFailed': 'Không thể lấy cấu hình',
   'roleConfig.fetchModelsFailed': 'Không thể tải danh sách mô hình',
   'roleConfig.fetchPluginsFailed': 'Không thể tải danh sách plugin',
+  'roleConfig.cannotResumeAudio': 'Không thể tiếp tục phát âm thanh',
+  'roleConfig.selectVoiceFirst': 'Vui lòng chọn giọng nói trước',
+  'roleConfig.audioLoadTimeout': 'Tải âm thanh mất quá nhiều thời gian, vui lòng thử lại sau',
+  'roleConfig.cloneAudioPlayFailed': 'Phát âm thanh sao chép thất bại',
+  'roleConfig.cannotPlayCloneAudio': 'Không thể phát âm thanh sao chép',
+  'roleConfig.getCloneAudioFailed': 'Không thể lấy âm thanh sao chép',
+  'roleConfig.noPreviewAudio': 'Giọng nói này không có âm thanh xem trước',
+  'roleConfig.audioPlayFailed': 'Phát âm thanh thất bại',
+  'roleConfig.cannotPlayAudio': 'Không thể phát âm thanh',
+  'roleConfig.audioPlayError': 'Lỗi trong quá trình phát âm thanh',
 
   // Function management dialog text
   'functionDialog.title': 'Quản lý chức năng',
@@ -919,6 +929,10 @@ export default {
   'ttsModel.selectVoiceToDelete': 'Vui lòng chọn giọng nói để xóa',
   'ttsModel.warning': 'Cảnh báo',
   'ttsModel.confirmDeleteVoice': 'Bạn có chắc chắn muốn xóa {count} giọng nói?',
+  'ttsModel.voiceCodeNameLanguageRequired': 'Mã giọng nói, tên giọng nói và loại ngôn ngữ không được để trống',
+  'ttsModel.deleteCancelled': 'Xóa đã bị hủy',
+  'ttsModel.updateSuccess': 'Cập nhật thành công',
+  'ttsModel.saveSuccess': 'Lưu thành công',
 
   // OTA Management Page Text
   'otaManagement.firmwareManagement': 'Quản lý firmware',

main/manager-web/src/i18n/zh_CN.js

@@ -779,6 +779,16 @@ export default {
   'roleConfig.fetchConfigFailed': '获取配置失败',
   'roleConfig.fetchModelsFailed': '获取模型列表失败',
   'roleConfig.fetchPluginsFailed': '获取插件列表失败',
+  'roleConfig.cannotResumeAudio': '无法恢复播放音频',
+  'roleConfig.selectVoiceFirst': '请先选择一个音色',
+  'roleConfig.audioLoadTimeout': '音频加载时间较长，请稍后重试',
+  'roleConfig.cloneAudioPlayFailed': '克隆音频播放失败',
+  'roleConfig.cannotPlayCloneAudio': '无法播放克隆音频',
+  'roleConfig.getCloneAudioFailed': '获取克隆音频失败',
+  'roleConfig.noPreviewAudio': '该音色暂无可预览的音频',
+  'roleConfig.audioPlayFailed': '音频播放失败',
+  'roleConfig.cannotPlayAudio': '无法播放音频',
+  'roleConfig.audioPlayError': '播放音频过程出错',
 
   // 功能管理对话框文本
   'functionDialog.title': '功能管理',
@@ -919,6 +929,10 @@ export default {
   'ttsModel.selectVoiceToDelete': '请选择要删除的音色',
   'ttsModel.warning': '警告',
   'ttsModel.confirmDeleteVoice': '确定要删除{count}个音色吗？',
+  'ttsModel.voiceCodeNameLanguageRequired': '音色编码、音色名称和语言类型不能为空',
+  'ttsModel.deleteCancelled': '删除已取消',
+  'ttsModel.updateSuccess': '更新成功',
+  'ttsModel.saveSuccess': '保存成功',
 
   // OTA管理页面文本
   'otaManagement.firmwareManagement': '固件管理',

main/manager-web/src/i18n/zh_TW.js

@@ -779,6 +779,16 @@ export default {
   'roleConfig.fetchConfigFailed': '獲取配置失敗',
   'roleConfig.fetchModelsFailed': '獲取模型列表失敗',
   'roleConfig.fetchPluginsFailed': '獲取插件列表失敗',
+  'roleConfig.cannotResumeAudio': '無法恢復播放音訊',
+  'roleConfig.selectVoiceFirst': '請先選擇一個音色',
+  'roleConfig.audioLoadTimeout': '音訊載入時間較長，請稍後重試',
+  'roleConfig.cloneAudioPlayFailed': '克隆音訊播放失敗',
+  'roleConfig.cannotPlayCloneAudio': '無法播放克隆音訊',
+  'roleConfig.getCloneAudioFailed': '獲取克隆音訊失敗',
+  'roleConfig.noPreviewAudio': '該音色暫無可預覽的音訊',
+  'roleConfig.audioPlayFailed': '音訊播放失敗',
+  'roleConfig.cannotPlayAudio': '無法播放音訊',
+  'roleConfig.audioPlayError': '播放音訊過程出錯',
 
   // 功能管理對話框文本
   'functionDialog.title': '功能管理',
@@ -917,6 +927,10 @@ export default {
   'ttsModel.selectVoiceToDelete': '請選擇要刪除的音色',
   'ttsModel.warning': '警告',
   'ttsModel.confirmDeleteVoice': '確定要刪除{count}個音色嗎？',
+  'ttsModel.voiceCodeNameLanguageRequired': '音色編碼、音色名稱和語言類型不能為空',
+  'ttsModel.deleteCancelled': '刪除已取消',
+  'ttsModel.updateSuccess': '更新成功',
+  'ttsModel.saveSuccess': '保存成功',
   'ttsModel.operationFailed': '操作失敗',
   'ttsModel.operationClosed': '操作已關閉',
 

main/manager-web/src/views/roleConfig.vue

@@ -599,21 +599,15 @@ export default {
           this.voiceOptions = data.data.map((voice) => ({
             value: voice.id,
             label: voice.name,
-            // 复制音频相关字段，确保hasAudioPreview能检测到
+            // 只保留后端实际返回的音频相关字段
             voiceDemo: voice.voiceDemo,
-            demoUrl: voice.demoUrl,
-            audioUrl: voice.audioUrl,
             voice_demo: voice.voice_demo,
-            sample_voice: voice.sample_voice,
-            referenceAudio: voice.referenceAudio,
-            // 新增：添加克隆音频相关字段
-            cloneAudioUrl: voice.cloneAudioUrl,
-            hasCloneAudio: voice.hasCloneAudio || false,
-            // 保存训练状态字段，用于判断是否为克隆音频
+            // 使用后端实际返回的 isClone 字段
+            isClone: Boolean(voice.isClone),
+            // 保存训练状态字段
             train_status: voice.trainStatus,
           }));
           // 保存完整的音色信息，添加调试信息
-          console.log("获取到的音色数据:", data.data);
           this.voiceDetails = data.data.reduce((acc, voice) => {
             acc[voice.id] = voice;
             return acc;
@@ -746,32 +740,15 @@ export default {
     },
     // 检查是否有音频预览
     hasAudioPreview(item) {
-      // 检查item中是否包含有效的音频URL字段或克隆音频字段
-      // 克隆音频通过hasCloneAudio标志或ID格式判断（非TTS开头的ID）
-      const isCloneAudio =
-        item.hasCloneAudio || (item.value && !item.value.startsWith("TTS"));
-
-      const audioFields = [
-        item.voiceDemo,
-        item.demoUrl,
-        item.audioUrl,
-        item.voice_demo,
-        item.sample_voice,
-        item.referenceAudio,
-        item.cloneAudioUrl, // 克隆音频的URL
-      ];
-
-      // 检查是否有任何音频字段是有效的URL
-      const hasUrlAudio = audioFields.some(
-        (field) =>
-          field !== undefined &&
-          field !== null &&
-          typeof field === "string" &&
-          field.trim() !== "" &&
-          field.toLowerCase().startsWith("http")
-      );
-
-      return hasUrlAudio || isCloneAudio;
+      // 检查是否为克隆音频
+      // 使用后端实际返回的 isClone 字段
+      const isCloneAudio = Boolean(item.isClone);
+      
+      // 检查是否有有效的音频URL，只使用后端实际返回的字段
+      const hasValidAudioUrl = !!((item.voice_demo || item.voiceDemo)?.trim());
+      
+      // 克隆音频始终显示播放按钮，普通音频需要有有效URL才显示
+      return isCloneAudio || hasValidAudioUrl;
     },
 
     // 播放/暂停音频切换
@@ -782,7 +759,7 @@ export default {
           // 从暂停状态恢复播放
           this.currentAudio.play().catch((error) => {
             console.error("恢复播放失败:", error);
-            this.$message.warning("无法恢复播放音频");
+            this.$message.warning(this.$t('roleConfig.cannotResumeAudio'));
           });
           this.isPaused = false;
         } else {
@@ -803,7 +780,7 @@ export default {
       const targetVoiceId = voiceId || this.form.ttsVoiceId;
 
       if (!targetVoiceId) {
-        this.$message.warning("请先选择一个音色");
+        this.$message.warning(this.$t('roleConfig.selectVoiceFirst'));
         return;
       }
 
@@ -830,10 +807,8 @@ export default {
         let isCloneAudio = false;
 
         if (voiceDetail) {
-          // 首先检查是否是克隆音频（通过ID格式判断，非TTS开头的ID）
-          isCloneAudio =
-            voiceDetail.hasCloneAudio ||
-            (voiceDetail.id && !voiceDetail.id.startsWith("TTS"));
+          // 使用后端实际返回的 isClone 字段判断是否为克隆音频
+          isCloneAudio = Boolean(voiceDetail.isClone);
           console.log(
             "克隆音频判断结果:",
             isCloneAudio,
@@ -889,7 +864,7 @@ export default {
             // 设置超时，防止加载过长时间
             const timeoutId = setTimeout(() => {
               if (this.currentAudio && this.playingVoice) {
-                this.$message.warning("音频加载时间较长，请稍后重试");
+                this.$message.warning(this.$t('roleConfig.audioLoadTimeout'));
                 this.playingVoice = false;
               }
             }, 10000); // 10秒超时
@@ -898,7 +873,7 @@ export default {
             this.currentAudio.onerror = () => {
               clearTimeout(timeoutId);
               console.error("克隆音频播放错误");
-              this.$message.warning("克隆音频播放失败");
+              this.$message.warning(this.$t('roleConfig.cloneAudioPlayFailed'));
               this.playingVoice = false;
             };
 
@@ -920,12 +895,12 @@ export default {
                 this.currentAudio.play().catch((error) => {
                   clearTimeout(timeoutId);
                   console.error("播放克隆音频失败:", error);
-                  this.$message.warning("无法播放克隆音频");
+                  this.$message.warning(this.$t('roleConfig.cannotPlayCloneAudio'));
                   this.playingVoice = false;
                 });
               } else {
                 clearTimeout(timeoutId);
-                this.$message.warning("获取克隆音频失败");
+                this.$message.warning(this.$t('roleConfig.getCloneAudioFailed'));
                 this.playingVoice = false;
               }
             });
@@ -933,14 +908,10 @@ export default {
             // 返回，避免继续执行下面的普通音频播放逻辑
             return;
           } else {
-            // 对于普通音频，尝试各种可能的URL字段
+            // 对于普通音频，只使用后端实际返回的字段
             audioUrl =
               voiceDetail.voiceDemo ||
-              voiceDetail.demoUrl ||
-              voiceDetail.audioUrl ||
-              voiceDetail.voice_demo ||
-              voiceDetail.sample_voice ||
-              voiceDetail.cloneAudioUrl; // 克隆音频URL
+              voiceDetail.voice_demo;
           }
 
           // 如果没有找到，尝试检查是否有URL格式的字段
@@ -965,7 +936,7 @@ export default {
 
         if (!audioUrl) {
           // 如果没有音频URL，显示友好的提示
-          this.$message.warning("该音色暂无可预览的音频");
+          this.$message.warning(this.$t('roleConfig.noPreviewAudio'));
           return;
         }
 
@@ -984,7 +955,7 @@ export default {
           // 设置超时，防止加载过长时间
           const timeoutId = setTimeout(() => {
             if (this.currentAudio && this.playingVoice) {
-              this.$message.warning("音频加载时间较长，请稍后重试");
+              this.$message.warning(this.$t('roleConfig.audioLoadTimeout'));
               this.playingVoice = false;
             }
           }, 10000); // 10秒超时
@@ -993,7 +964,7 @@ export default {
           this.currentAudio.onerror = () => {
             clearTimeout(timeoutId);
             console.error("音频播放错误");
-            this.$message.warning("音频播放失败");
+            this.$message.warning(this.$t('roleConfig.audioPlayFailed'));
             this.playingVoice = false;
           };
 
@@ -1011,13 +982,13 @@ export default {
           this.currentAudio.play().catch((error) => {
             clearTimeout(timeoutId);
             console.error("播放失败:", error);
-            this.$message.warning("无法播放音频");
+            this.$message.warning(this.$t('roleConfig.cannotPlayAudio'));
             this.playingVoice = false;
           });
         }
       } catch (error) {
         console.error("播放音频过程出错:", error);
-        this.$message.error("播放音频过程出错");
+        this.$message.error(this.$t('roleConfig.audioPlayError'));
         this.playingVoice = false;
       }
     },

main/xiaozhi-server/core/handle/helloHandle.py

@@ -1,5 +1,6 @@
 import time
 import json
+import uuid
 import random
 import asyncio
 from core.utils.dialogue import Message
@@ -105,6 +106,9 @@ async def checkWakeupWords(conn, text):
     # 播放唤醒词回复
     conn.client_abort = False
 
+    # 将唤醒词回复视为新会话，生成新的 sentence_id，确保流控器重置
+    conn.sentence_id = str(uuid.uuid4().hex)
+
     conn.logger.bind(tag=TAG).info(f"播放唤醒词回复: {response.get('text')}")
     await sendAudioMessage(conn, SentenceType.FIRST, opus_packets, response.get("text"))
     await sendAudioMessage(conn, SentenceType.LAST, [], None)

main/xiaozhi-server/core/handle/sendAudioHandle.py

@@ -133,12 +133,27 @@ def _get_or_create_rate_controller(conn, frame_duration, is_single_packet):
     Returns:
         (rate_controller, flow_control)
     """
-    # 判断是否需要重置：单包模式且 sentence_id 变化，或者控制器不存在
-    need_reset = (
-        is_single_packet
-        and getattr(conn, "audio_flow_control", {}).get("sentence_id")
-        != conn.sentence_id
-    ) or not hasattr(conn, "audio_rate_controller")
+    # 检查是否需要重置控制器
+    need_reset = False
+
+    if not hasattr(conn, "audio_rate_controller"):
+        # 控制器不存在，需要创建
+        need_reset = True
+    else:
+        rate_controller = conn.audio_rate_controller
+
+        # 后台发送任务已停止, 则需要重置
+        if (
+            not rate_controller.pending_send_task
+            or rate_controller.pending_send_task.done()
+        ):
+            need_reset = True
+        # 当sentence_id 变化，需要重置
+        elif (
+            getattr(conn, "audio_flow_control", {}).get("sentence_id")
+            != conn.sentence_id
+        ):
+            need_reset = True
 
     if need_reset:
         # 创建或获取 rate_controller

main/xiaozhi-server/core/providers/memory/powermem/powermem.py

@@ -46,7 +46,7 @@ class MemoryProvider(MemoryProviderBase):
         try:
             # Check if user profile mode is enabled
             self.enable_user_profile = config.get("enable_user_profile", False)
-
+            
             # Get configuration parameters
             database_provider = config.get("database_provider", "sqlite")
             llm_provider = config.get("llm_provider", "qwen")
@@ -134,12 +134,12 @@ class MemoryProvider(MemoryProviderBase):
                 memory_mode = "AsyncMemory (普通记忆模式)"
 
             self.use_powermem = True
-
+            
             logger.bind(tag=TAG).info(
                 f"PowerMem initialized successfully: mode={memory_mode}, "
                 f"database={database_provider}, llm={llm_provider}, embedding={embedding_provider}"
             )
-
+            
         except ImportError as e:
             logger.bind(tag=TAG).error(
                 f"PowerMem not installed. Please install with: pip install powermem. Error: {e}"
@@ -150,13 +150,15 @@ class MemoryProvider(MemoryProviderBase):
             logger.bind(tag=TAG).debug(f"Detailed error: {traceback.format_exc()}")
             self.use_powermem = False
 
-    async def save_memory(self, msgs):
+    async def save_memory(self, msgs, session_id=None):
         """
         Save conversation messages to PowerMem.
 
         Args:
             msgs: List of message objects with 'role' and 'content' attributes
 
+            session_id: Session identifier (optional, for compatibility)
+
         Returns:
             Result from PowerMem API or None if failed
         """
@@ -186,13 +188,13 @@ class MemoryProvider(MemoryProviderBase):
                 result = await result
 
             logger.bind(tag=TAG).debug(f"Save memory result: {result}")
-            
+
             # Cache user profile if UserMemory mode and profile was extracted
             if self.enable_user_profile and result:
                 if result.get('profile_extracted'):
                     self.last_profile_content = result.get('profile_content', '')
                     logger.bind(tag=TAG).debug(f"User profile extracted: {self.last_profile_content}")
-            
+
             return result
 
         except Exception as e:
@@ -306,3 +308,6 @@ class MemoryProvider(MemoryProviderBase):
 
         return ""
 
+
+# Register the memory provider instance
+powermem = MemoryProvider({})

main/xiaozhi-server/core/providers/tts/alibl_stream.py

@@ -185,20 +185,21 @@ class TTSProvider(TTSProviderBase):
             # 过滤Markdown
             filtered_text = MarkdownCleaner.clean_markdown(text)
 
-            # 发送continue-task消息
-            continue_task_message = {
-                "header": {
-                    "action": "continue-task",
-                    "task_id": self.conn.sentence_id,
-                    "streaming": "duplex",
-                },
-                "payload": {"input": {"text": filtered_text}},
-            }
-
-            await self.ws.send(json.dumps(continue_task_message))
-            self.last_active_time = time.time()
-            logger.bind(tag=TAG).debug(f"已发送文本: {filtered_text}")
+            if filtered_text:
+                # 发送continue-task消息
+                continue_task_message = {
+                    "header": {
+                        "action": "continue-task",
+                        "task_id": self.conn.sentence_id,
+                        "streaming": "duplex",
+                    },
+                    "payload": {"input": {"text": filtered_text}},
+                }
 
+                await self.ws.send(json.dumps(continue_task_message))
+                self.last_active_time = time.time()
+                logger.bind(tag=TAG).debug(f"已发送文本: {filtered_text}")
+            return
         except Exception as e:
             logger.bind(tag=TAG).error(f"发送TTS文本失败: {str(e)}")
             if self.ws:

main/xiaozhi-server/core/providers/tts/aliyun_stream.py

@@ -288,18 +288,19 @@ class TTSProvider(TTSProviderBase):
                 logger.bind(tag=TAG).warning(f"WebSocket连接不存在，终止发送文本")
                 return
             filtered_text = MarkdownCleaner.clean_markdown(text)
-            run_request = {
-                "header": {
-                    "message_id": uuid.uuid4().hex,
-                    "task_id": self.task_id,
-                    "namespace": "FlowingSpeechSynthesizer",
-                    "name": "RunSynthesis",
-                    "appkey": self.appkey,
-                },
-                "payload": {"text": filtered_text},
-            }
-            await self.ws.send(json.dumps(run_request))
-            self.last_active_time = time.time()
+            if filtered_text:
+                run_request = {
+                    "header": {
+                        "message_id": uuid.uuid4().hex,
+                        "task_id": self.task_id,
+                        "namespace": "FlowingSpeechSynthesizer",
+                        "name": "RunSynthesis",
+                        "appkey": self.appkey,
+                    },
+                    "payload": {"text": filtered_text},
+                }
+                await self.ws.send(json.dumps(run_request))
+                self.last_active_time = time.time()
             return
 
         except Exception as e:

main/xiaozhi-server/core/providers/tts/huoshan_double_stream.py

@@ -4,15 +4,15 @@ import json
 import queue
 import asyncio
 import traceback
-from typing import Callable, Any
 import websockets
+
+from typing import Callable, Any
 from core.utils.tts import MarkdownCleaner
 from config.logger import setup_logging
 from core.utils import opus_encoder_utils
 from core.utils.util import check_model_key
 from core.providers.tts.base import TTSProviderBase
 from core.providers.tts.dto.dto import SentenceType, ContentType, InterfaceType
-from asyncio import Task
 
 
 TAG = __name__
@@ -340,8 +340,9 @@ class TTSProvider(TTSProviderBase):
             #  过滤Markdown
             filtered_text = MarkdownCleaner.clean_markdown(text)
 
-            # 发送文本
-            await self.send_text(self.voice, filtered_text, self.conn.sentence_id)
+            if filtered_text:
+                # 发送文本
+                await self.send_text(self.voice, filtered_text, self.conn.sentence_id)
             return
         except Exception as e:
             logger.bind(tag=TAG).error(f"发送TTS文本失败: {str(e)}")

main/xiaozhi-server/core/providers/tts/xunfei_stream.py

@@ -237,10 +237,10 @@ class TTSProvider(TTSProviderBase):
                 return
 
             filtered_text = MarkdownCleaner.clean_markdown(text)
-
-            # 发送文本合成请求
-            run_request = self._build_base_request(status=1,text=filtered_text)
-            await self.ws.send(json.dumps(run_request))
+            if filtered_text:
+                # 发送文本合成请求
+                run_request = self._build_base_request(status=1,text=filtered_text)
+                await self.ws.send(json.dumps(run_request))
             return
 
         except Exception as e:

main/xiaozhi-server/core/utils/textUtils.py

@@ -2,7 +2,7 @@ import json
 
 TAG = __name__
 EMOJI_MAP = {
-    "😂": "laughing",
+    "😂": "funny",
     "😭": "crying",
     "😠": "angry",
     "😔": "sad",

main/xiaozhi-server/core/utils/tts.py

@@ -1,9 +1,11 @@
 import os
 import re
 import sys
-from config.logger import setup_logging
 import importlib
 
+from config.logger import setup_logging
+from core.utils.textUtils import check_emoji
+
 logger = setup_logging()
 
 punctuation_set = {
@@ -135,4 +137,8 @@ class MarkdownCleaner:
 
         for regex, replacement in MarkdownCleaner.REGEXES:
             text = regex.sub(replacement, text)
+
+        # 去除emoji表情
+        text = check_emoji(text)
+
         return text.strip()