TTS文本过滤

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
+            if filtered_text:
-            continue_task_message = {
+                # 发送continue-task消息
-                "header": {
+                continue_task_message = {
-                    "action": "continue-task",
+                    "header": {
-                    "task_id": self.conn.sentence_id,
+                        "action": "continue-task",
-                    "streaming": "duplex",
+                        "task_id": self.conn.sentence_id,
-                },
+                        "streaming": "duplex",
-                "payload": {"input": {"text": filtered_text}},
+                    },
-            }
+                    "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}")
 
+                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 = {
+            if filtered_text:
-                "header": {
+                run_request = {
-                    "message_id": uuid.uuid4().hex,
+                    "header": {
-                    "task_id": self.task_id,
+                        "message_id": uuid.uuid4().hex,
-                    "namespace": "FlowingSpeechSynthesizer",
+                        "task_id": self.task_id,
-                    "name": "RunSynthesis",
+                        "namespace": "FlowingSpeechSynthesizer",
-                    "appkey": self.appkey,
+                        "name": "RunSynthesis",
-                },
+                        "appkey": self.appkey,
-                "payload": {"text": filtered_text},
+                    },
-            }
+                    "payload": {"text": filtered_text},
-            await self.ws.send(json.dumps(run_request))
+                }
-            self.last_active_time = time.time()
+                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
+            if filtered_text:
-            await self.send_text(self.voice, filtered_text, self.conn.sentence_id)
+                # 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)
-
+            if filtered_text:
-            # 发送文本合成请求
+                # 发送文本合成请求
-            run_request = self._build_base_request(status=1,text=filtered_text)
+                run_request = self._build_base_request(status=1,text=filtered_text)
-            await self.ws.send(json.dumps(run_request))
+                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()