refactor: 优化超时配置，响应社区反馈

- 工具调用超时改为可配置项 tool_call_timeout，默认30秒
- OpenAI 超时恢复可配置机制，支持细粒度配置和单一值
- TTS 超时保护统一添加到所有流式TTS实现
- 将 ActionResponse 导入移到文件顶部
- 修复超时配置边界情况处理

main/xiaozhi-server/config.yaml

@@ -61,6 +61,8 @@ delete_audio: true
 close_connection_no_voice_time: 120
 # TTS请求超时时间(秒)
 tts_timeout: 10
+# 工具调用超时时间(秒)
+tool_call_timeout: 30
 # 开启唤醒词加速
 enable_wakeup_words_response_cache: true
 # 开场是否回复唤醒词

main/xiaozhi-server/core/connection.py

@@ -32,7 +32,7 @@ from core.providers.asr.dto.dto import InterfaceType
 from core.handle.textHandle import handleTextMessage
 from core.providers.tools.unified_tool_handler import UnifiedToolHandler
 from plugins_func.loadplugins import auto_import_modules
-from plugins_func.register import Action
+from plugins_func.register import Action, ActionResponse
 from core.auth import AuthenticationError
 from config.config_loader import get_private_config_from_api
 from core.providers.tts.dto.dto import ContentType, TTSMessageDTO, SentenceType
@@ -997,19 +997,19 @@ class ConnectionHandler:
                     )
                     futures_with_data.append((future, tool_call_data))
 
-                TOOL_CALL_TIMEOUT = 30
+                # 工具调用超时时间，可配置，默认30秒
+                tool_call_timeout = int(self.config.get("tool_call_timeout", 30))
                 # 等待协程结束（实际等待时长为最慢的那个）
                 tool_results = []
                 for future, tool_call_data in futures_with_data:
                     try:
-                        result = future.result(timeout=TOOL_CALL_TIMEOUT)
+                        result = future.result(timeout=tool_call_timeout)
                         tool_results.append((result, tool_call_data))
                     except Exception as e:
                         self.logger.bind(tag=TAG).error(
                             f"工具调用超时或异常: {tool_call_data['name']}, 错误: {e}"
                         )
                         # 超时时返回错误响应，避免整个流程卡死
-                        from plugins_func.register import Action, ActionResponse
                         tool_results.append((
                             ActionResponse(action=Action.ERROR, result="哎呀，网络遇到点问题，请稍后再试下！"),
                             tool_call_data

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

@@ -150,8 +150,8 @@ async def process_intent_result(
             def process_function_call():
                 conn.dialogue.put(Message(role="user", content=original_text))
                 
-                # 使用统一工具处理器处理所有工具调用，添加超时保护
+                # 工具调用超时时间
-                TOOL_CALL_TIMEOUT = 30
+                tool_call_timeout = int(conn.config.get("tool_call_timeout", 30))
                 # 使用统一工具处理器处理所有工具调用
                 try:
                     result = asyncio.run_coroutine_threadsafe(
@@ -159,7 +159,7 @@ async def process_intent_result(
                             conn, function_call_data
                         ),
                         conn.loop,
-                    ).result(timeout=TOOL_CALL_TIMEOUT)
+                    ).result(timeout=tool_call_timeout)
                 except Exception as e:
                     conn.logger.bind(tag=TAG).error(f"工具调用失败: {e}")
                     result = ActionResponse(

main/xiaozhi-server/core/providers/llm/openai/openai.py

@@ -17,12 +17,22 @@ class LLMProvider(LLMProviderBase):
             self.base_url = config.get("base_url")
         else:
             self.base_url = config.get("url")
-        custom_timeout = httpx.Timeout(
+        
-            pool=2.0,
+        timeout_config = config.get("timeout")
-            connect=3.0,
+        if isinstance(timeout_config, dict):
-            write=5.0,
+            # 细粒度超时配置
-            read=10.0
+            custom_timeout = httpx.Timeout(
-        )
+                pool=timeout_config.get("pool", 2.0),
+                connect=timeout_config.get("connect", 3.0),
+                write=timeout_config.get("write", 5.0),
+                read=timeout_config.get("read", 60.0)
+            )
+        elif isinstance(timeout_config, (int, float)) and timeout_config > 0:
+            # 兼容旧的单一超时配置（整数或浮点数）
+            custom_timeout = httpx.Timeout(timeout_config)
+        else:
+            # 未配置或配置无效，使用默认值
+            custom_timeout = httpx.Timeout(300)
 
         param_defaults = {
             "max_tokens": int,

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

@@ -127,7 +127,7 @@ class TTSProvider(TTSProviderBase):
                             self.start_session(self.conn.sentence_id),
                             loop=self.conn.loop,
                         )
-                        future.result()
+                        future.result(timeout=self.tts_timeout)
                         self.before_stop_play_files.clear()
                         logger.bind(tag=TAG).info("TTS会话启动成功")
                     except Exception as e:
@@ -144,7 +144,7 @@ class TTSProvider(TTSProviderBase):
                                 self.text_to_speak(message.content_detail, None),
                                 loop=self.conn.loop,
                             )
-                            future.result()
+                            future.result(timeout=self.tts_timeout)
                             logger.bind(tag=TAG).debug("TTS文本发送成功")
                         except Exception as e:
                             logger.bind(tag=TAG).error(f"发送TTS文本失败: {str(e)}")
@@ -165,7 +165,7 @@ class TTSProvider(TTSProviderBase):
                             self.finish_session(self.conn.sentence_id),
                             loop=self.conn.loop,
                         )
-                        future.result()
+                        future.result(timeout=self.tts_timeout)
                     except Exception as e:
                         logger.bind(tag=TAG).error(f"结束TTS会话失败: {str(e)}")
                         continue

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

@@ -232,7 +232,7 @@ class TTSProvider(TTSProviderBase):
                             self.start_session(self.task_id),
                             loop=self.conn.loop,
                         )
-                        future.result()
+                        future.result(timeout=self.tts_timeout)
                         self.before_stop_play_files.clear()
                         logger.bind(tag=TAG).debug("TTS会话启动成功")
 
@@ -250,7 +250,7 @@ class TTSProvider(TTSProviderBase):
                                 self.text_to_speak(message.content_detail, None),
                                 loop=self.conn.loop,
                             )
-                            future.result()
+                            future.result(timeout=self.tts_timeout)
                             logger.bind(tag=TAG).debug("TTS文本发送成功")
                         except Exception as e:
                             logger.bind(tag=TAG).error(f"发送TTS文本失败: {str(e)}")
@@ -270,7 +270,7 @@ class TTSProvider(TTSProviderBase):
                             self.finish_session(self.task_id),
                             loop=self.conn.loop,
                         )
-                        future.result()
+                        future.result(timeout=self.tts_timeout)
                     except Exception as e:
                         logger.bind(tag=TAG).error(f"结束TTS会话失败: {str(e)}")
                         continue

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

@@ -36,6 +36,7 @@ class TTSProviderBase(ABC):
         self.delete_audio_file = delete_audio_file
         self.audio_file_type = "wav"
         self.output_file = config.get("output_dir", "tmp/")
+        self.tts_timeout = int(config.get("tts_timeout", 15))
         self.tts_text_queue = queue.Queue()
         self.tts_audio_queue = queue.Queue()
         self.tts_audio_first_sentence = True
@@ -357,7 +358,7 @@ class TTSProviderBase(ABC):
                     sendAudioMessage(self.conn, sentence_type, audio_datas, text),
                     self.conn.loop,
                 )
-                future.result(timeout=15)
+                future.result(timeout=self.tts_timeout)
 
                 # 记录输出和报告
                 if self.conn.max_output_size > 0 and text:

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

@@ -149,7 +149,6 @@ class TTSProvider(TTSProviderBase):
         self.cluster = config.get("cluster")
         self.resource_id = config.get("resource_id")
         self.activate_session = False
-        self.timeout = 30
         if config.get("private_voice"):
             self.voice = config.get("private_voice")
         else:
@@ -308,7 +307,7 @@ class TTSProvider(TTSProviderBase):
                             self.start_session(self.conn.sentence_id),
                             loop=self.conn.loop,
                         )
-                        future.result(self.timeout)
+                        future.result(timeout=self.tts_timeout)
                         self.before_stop_play_files.clear()
                         logger.bind(tag=TAG).debug("TTS会话启动成功")
                     except Exception as e:
@@ -325,7 +324,7 @@ class TTSProvider(TTSProviderBase):
                                 self.text_to_speak(message.content_detail, None),
                                 loop=self.conn.loop,
                             )
-                            future.result(self.timeout)
+                            future.result(timeout=self.tts_timeout)
                             logger.bind(tag=TAG).debug("TTS文本发送成功")
                         except Exception as e:
                             logger.bind(tag=TAG).error(f"发送TTS文本失败: {str(e)}")
@@ -345,7 +344,7 @@ class TTSProvider(TTSProviderBase):
                             self.finish_session(self.conn.sentence_id),
                             loop=self.conn.loop,
                         )
-                        future.result(self.timeout)
+                        future.result(timeout=self.tts_timeout)
                     except Exception as e:
                         logger.bind(tag=TAG).error(f"结束TTS会话失败: {str(e)}")
                         continue

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

@@ -178,7 +178,7 @@ class TTSProvider(TTSProviderBase):
                             self.start_session(self.conn.sentence_id),
                             loop=self.conn.loop,
                         )
-                        future.result()
+                        future.result(timeout=self.tts_timeout)
                         self.before_stop_play_files.clear()
                         logger.bind(tag=TAG).info("TTS会话启动成功")
 
@@ -197,7 +197,7 @@ class TTSProvider(TTSProviderBase):
                                 self.text_to_speak(message.content_detail, None),
                                 loop=self.conn.loop,
                             )
-                            future.result()
+                            future.result(timeout=self.tts_timeout)
                             logger.bind(tag=TAG).debug("TTS文本发送成功")
                         except Exception as e:
                             logger.bind(tag=TAG).error(f"发送TTS文本失败: {str(e)}")