Merge branch 'main' into main

config.yaml

@@ -52,7 +52,7 @@ VAD:
     min_silence_duration_ms: 700  # 如果说话停顿比较长，可以把这个值设置大一些
 
 LLM:
-  # 当前支持的type为openai、dify，可自行适配
+  # 当前支持的type为openai、dify、ollama，可自行适配
   AliLLM:
     # 定义LLM API类型
     type: openai
@@ -75,6 +75,11 @@ LLM:
     model_name: glm-4-flash
     url: https://open.bigmodel.cn/api/paas/v4/
     api_key: 你的ChatGLMLLM api key
+  OllamaLLM:
+    # 定义LLM API类型
+    type: ollama
+    model_name: qwen2.5 #  使用的模型名称，需要预先使用ollama pull下载
+    base_url: http://localhost:11434  # Ollama服务地址
   DifyLLM:
     # 定义LLM API类型
     type: dify

core/connection.py

@@ -155,9 +155,10 @@ class ConnectionHandler:
         # 处理剩余的响应
         if start < len(response_message):
             segment_text = "".join(response_message[start:])
-            self.recode_first_last_text(segment_text)
-            future = self.executor.submit(self.speak_and_play, segment_text)
-            self.tts_queue.put(future)
+            if len(segment_text) > 0:
+                self.recode_first_last_text(segment_text)
+                future = self.executor.submit(self.speak_and_play, segment_text)
+                self.tts_queue.put(future)
 
         self.llm_finish_task = True
         # 更新对话
@@ -176,6 +177,11 @@ class ConnectionHandler:
                 try:
                     self.logger.debug("正在处理TTS任务...")
                     tts_file, text = future.result(timeout=10)
+                    if text is None or len(text) <= 0:
+                        continue
+                    if tts_file is None:
+                        self.logger.error(f"TTS文件生成失败: {text}")
+                        continue
                     self.logger.debug(f"TTS文件生成完毕，文件路径: {tts_file}")
                     if os.path.exists(tts_file):
                         opus_datas, duration = self.tts.wav_to_opus_data(tts_file)
@@ -208,11 +214,11 @@ class ConnectionHandler:
     def speak_and_play(self, text):
         if text is None or len(text) <= 0:
             self.logger.info(f"无需tts转换，query为空，{text}")
-            return None
+            return None, text
         tts_file = self.tts.to_tts(text)
         if tts_file is None:
             self.logger.error(f"tts转换失败，{text}")
-            return None
+            return None, text
         self.logger.debug(f"TTS 文件生成完毕: {tts_file}")
         return tts_file, text
 

core/providers/llm/ollama/ollama.py

@@ -0,0 +1,45 @@
+import logging
+import requests, json
+from core.providers.llm.base import LLMProviderBase
+
+logger = logging.getLogger(__name__)
+
+
+class LLMProvider(LLMProviderBase):
+    def __init__(self, config):
+
+        self.model_name = config.get("model_name")
+        self.base_url = config.get("base_url", "http://localhost:11434")
+
+    def response(self, session_id, dialogue):        
+        try:
+            # Convert dialogue format to Ollama format
+            prompt = ""
+            for msg in dialogue:
+                if msg["role"] == "system":
+                    prompt += f"System: {msg['content']}\n"
+                elif msg["role"] == "user":
+                    prompt += f"User: {msg['content']}\n"
+                elif msg["role"] == "assistant":
+                    prompt += f"Assistant: {msg['content']}\n"
+
+            # Make request to Ollama API
+            response = requests.post(
+                f"{self.base_url}/api/generate",
+                json={
+                    "model": self.model_name,
+                    "prompt": prompt,
+                    "stream": True
+                },
+                stream=True
+            )
+
+            for line in response.iter_lines():
+                if line:
+                    json_response = json.loads(line)
+                    if "response" in json_response:
+                        yield json_response["response"]
+
+        except Exception as e:
+            logger.error(f"Error in Ollama response generation: {e}")
+            yield "【Ollama服务响应异常】"