Hass/xiaozhi-esp32-server
1
0
Fork 0
mirror of https://github.com/xinnan-tech/xiaozhi-esp32-server.git synced 2026-07-27 01:23:55 +08:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #1564 from xinnan-tech/agent-plugin

Browse Source 
Agent plugin
This commit is contained in:
欣南科技
2025-06-13 09:48:38 +08:00
committed by GitHub
parent 25ccf7fed3 8a2fe7c20e
commit 2ff7fbbe6f
54 changed files with 1469 additions and 699 deletions
Download Patch File Download Diff File Expand all files Collapse all files
main/xiaozhi-server/core/connection.py
+11 -3
View File
@@ -453,9 +453,17 @@ class ConnectionHandler:
if private_config.get("Intent", None) is not None:
init_intent = True
self.config["Intent"] = private_config["Intent"]
self.config["selected_module"]["Intent"] = private_config[
"selected_module"
]["Intent"]
model_intent = private_config.get("selected_module", {}).get("Intent", {})
self.config["selected_module"]["Intent"] = model_intent
# 加载插件配置
if model_intent != "Intent_nointent":
plugin_from_server = private_config.get("plugins", {})
for plugin, config_str in plugin_from_server.items():
plugin_from_server[plugin] = json.loads(config_str)
self.config["plugins"] = plugin_from_server
self.config["Intent"][self.config["selected_module"]["Intent"]][
"functions"
] = plugin_from_server.keys()
if private_config.get("prompt", None) is not None:
self.config["prompt"] = private_config["prompt"]
if private_config.get("summaryMemory", None) is not None:
main/xiaozhi-server/core/handle/functionHandler.py
-1
View File
@@ -61,7 +61,6 @@ class FunctionHandler:
self.function_registry.register_function("plugin_loader")
self.function_registry.register_function("get_time")
self.function_registry.register_function("get_lunar")
# self.function_registry.register_function("handle_speaker_volume_or_screen_brightness")
def register_config_functions(self):
"""注册配置中的函数,可以不同客户端使用不同的配置"""
main/xiaozhi-server/core/providers/llm/coze/coze.py
+3 -1
View File
@@ -23,7 +23,9 @@ class LLMProvider(LLMProviderBase):
self.bot_id = str(config.get("bot_id"))
self.user_id = str(config.get("user_id"))
self.session_conversation_map = {} # 存储session_id和conversation_id的映射
check_model_key("CozeLLM", self.personal_access_token)
model_key_msg = check_model_key("CozeLLM", self.personal_access_token)
if model_key_msg:
logger.bind(tag=TAG).error(model_key_msg)
def response(self, session_id, dialogue, **kwargs):
coze_api_token = self.personal_access_token
main/xiaozhi-server/core/providers/llm/dify/dify.py
+3 -1
View File
@@ -15,7 +15,9 @@ class LLMProvider(LLMProviderBase):
self.mode = config.get("mode", "chat-messages")
self.base_url = config.get("base_url", "https://api.dify.ai/v1").rstrip("/")
self.session_conversation_map = {} # 存储session_id和conversation_id的映射
check_model_key("DifyLLM", self.api_key)
model_key_msg = check_model_key("DifyLLM", self.api_key)
if model_key_msg:
logger.bind(tag=TAG).error(model_key_msg)
def response(self, session_id, dialogue, **kwargs):
try:
main/xiaozhi-server/core/providers/llm/fastgpt/fastgpt.py
+3 -1
View File
@@ -14,7 +14,9 @@ class LLMProvider(LLMProviderBase):
self.base_url = config.get("base_url")
self.detail = config.get("detail", False)
self.variables = config.get("variables", {})
check_model_key("FastGPTLLM", self.api_key)
model_key_msg = check_model_key("FastGPTLLM", self.api_key)
if model_key_msg:
logger.bind(tag=TAG).error(model_key_msg)
def response(self, session_id, dialogue, **kwargs):
try:
main/xiaozhi-server/core/providers/llm/gemini/gemini.py
+3 -2
View File
@@ -73,8 +73,9 @@ class LLMProvider(LLMProviderBase):
http_proxy = cfg.get("http_proxy")
https_proxy = cfg.get("https_proxy")
if not check_model_key("LLM", self.api_key):
raise ValueError("无效的Gemini API Key，请检查是否配置正确")
model_key_msg = check_model_key("LLM", self.api_key)
if model_key_msg:
log.bind(tag=TAG).error(model_key_msg)
if http_proxy or https_proxy:
log.bind(tag=TAG).info(
main/xiaozhi-server/core/providers/llm/openai/openai.py
+20 -9
View File
@@ -21,20 +21,27 @@ class LLMProvider(LLMProviderBase):
"max_tokens": (500, int),
"temperature": (0.7, lambda x: round(float(x), 1)),
"top_p": (1.0, lambda x: round(float(x), 1)),
"frequency_penalty": (0, lambda x: round(float(x), 1))
"frequency_penalty": (0, lambda x: round(float(x), 1)),
}
for param, (default, converter) in param_defaults.items():
value = config.get(param)
try:
setattr(self, param, converter(value) if value not in (None, "") else default)
setattr(
self,
param,
converter(value) if value not in (None, "") else default,
)
except (ValueError, TypeError):
setattr(self, param, default)
logger.debug(
f"意图识别参数初始化: {self.temperature}, {self.max_tokens}, {self.top_p}, {self.frequency_penalty}")
f"意图识别参数初始化: {self.temperature}, {self.max_tokens}, {self.top_p}, {self.frequency_penalty}"
)
check_model_key("LLM", self.api_key)
model_key_msg = check_model_key("LLM", self.api_key)
if model_key_msg:
logger.bind(tag=TAG).error(model_key_msg)
self.client = openai.OpenAI(api_key=self.api_key, base_url=self.base_url)
def response(self, session_id, dialogue, **kwargs):
@@ -46,7 +53,9 @@ class LLMProvider(LLMProviderBase):
max_tokens=kwargs.get("max_tokens", self.max_tokens),
temperature=kwargs.get("temperature", self.temperature),
top_p=kwargs.get("top_p", self.top_p),
frequency_penalty=kwargs.get("frequency_penalty", self.frequency_penalty),
frequency_penalty=kwargs.get(
"frequency_penalty", self.frequency_penalty
),
)
is_active = True
@@ -84,12 +93,14 @@ class LLMProvider(LLMProviderBase):
for chunk in stream:
# 检查是否存在有效的choice且content不为空
if getattr(chunk, "choices", None):
yield chunk.choices[0].delta.content, chunk.choices[0].delta.tool_calls
yield chunk.choices[0].delta.content, chunk.choices[
0
].delta.tool_calls
# 存在 CompletionUsage 消息时，生成 Token 消耗 log
elif isinstance(getattr(chunk, 'usage', None), CompletionUsage):
usage_info = getattr(chunk, 'usage', None)
elif isinstance(getattr(chunk, "usage", None), CompletionUsage):
usage_info = getattr(chunk, "usage", None)
logger.bind(tag=TAG).info(
f"Token 消耗：输入 {getattr(usage_info, 'prompt_tokens', '未知')}，"
f"Token 消耗：输入 {getattr(usage_info, 'prompt_tokens', '未知')}，"
f"输出 {getattr(usage_info, 'completion_tokens', '未知')}，"
f"共计 {getattr(usage_info, 'total_tokens', '未知')}"
)
main/xiaozhi-server/core/providers/memory/base.py
-4
View File
@@ -12,10 +12,6 @@ class MemoryProviderBase(ABC):
def set_llm(self, llm):
self.llm = llm
# 获取模型名称和类型信息
model_name = getattr(llm, "model_name", str(llm.__class__.__name__))
# 记录更详细的日志
logger.bind(tag=TAG).info(f"记忆总结设置LLM: {model_name}")
@abstractmethod
async def save_memory(self, msgs):
main/xiaozhi-server/core/providers/memory/mem0ai/mem0ai.py
+4 -2
View File
@@ -12,12 +12,14 @@ class MemoryProvider(MemoryProviderBase):
super().__init__(config)
self.api_key = config.get("api_key", "")
self.api_version = config.get("api_version", "v1.1")
have_key = check_model_key("Mem0ai", self.api_key)
if not have_key:
model_key_msg = check_model_key("Mem0ai", self.api_key)
if model_key_msg:
logger.bind(tag=TAG).error(model_key_msg)
self.use_mem0 = False
return
else:
self.use_mem0 = True
try:
self.client = MemoryClient(api_key=self.api_key)
logger.bind(tag=TAG).info("成功连接到 Mem0ai 服务")
main/xiaozhi-server/core/providers/tts/doubao.py
+3 -1
View File
@@ -37,7 +37,9 @@ class TTSProvider(TTSProviderBase):
self.api_url = config.get("api_url")
self.authorization = config.get("authorization")
self.header = {"Authorization": f"{self.authorization}{self.access_token}"}
check_model_key("TTS", self.access_token)
model_key_msg = check_model_key("TTS", self.access_token)
if model_key_msg:
logger.bind(tag=TAG).error(model_key_msg)
async def text_to_speak(self, text, output_file):
request_json = {
main/xiaozhi-server/core/providers/tts/fishspeech.py
+3 -2
View File
@@ -90,8 +90,9 @@ class TTSProvider(TTSProviderBase):
self.format = config.get("response_format", "wav")
self.audio_file_type = config.get("response_format", "wav")
self.api_key = config.get("api_key", "YOUR_API_KEY")
have_key = check_model_key("FishSpeech TTS", self.api_key)
if not have_key:
model_key_msg = check_model_key("FishSpeech TTS", self.api_key)
if model_key_msg:
logger.bind(tag=TAG).error(model_key_msg)
return
self.normalize = str(config.get("normalize", True)).lower() in (
"true",
main/xiaozhi-server/core/providers/tts/huoshan_double_stream.py
+4 -2
View File
@@ -157,7 +157,9 @@ class TTSProvider(TTSProviderBase):
self.opus_encoder = opus_encoder_utils.OpusEncoderUtils(
sample_rate=16000, channels=1, frame_size_ms=60
)
check_model_key("TTS", self.access_token)
model_key_msg = check_model_key("TTS", self.access_token)
if model_key_msg:
logger.bind(tag=TAG).error(model_key_msg)
async def open_audio_channels(self, conn):
try:
@@ -267,7 +269,7 @@ class TTSProvider(TTSProviderBase):
await handleAbortMessage(self.conn)
logger.bind(tag=TAG).error(f"WebSocket连接不存在，终止发送文本")
return
# 过滤Markdown
filtered_text = MarkdownCleaner.clean_markdown(text)
main/xiaozhi-server/core/providers/tts/linkerai.py
+6 -4
View File
@@ -166,7 +166,9 @@ class TTSProvider(TTSProviderBase):
) as resp:
if resp.status != 200:
logger.error(f"TTS请求失败: {resp.status}, {await resp.text()}")
logger.bind(tag=TAG).error(
f"TTS请求失败: {resp.status}, {await resp.text()}"
)
self.tts_audio_queue.put((SentenceType.LAST, [], None))
return
@@ -229,7 +231,7 @@ class TTSProvider(TTSProviderBase):
self._process_before_stop_play_files()
except Exception as e:
logger.error(f"TTS请求异常: {e}")
logger.bind(tag=TAG).error(f"TTS请求异常: {e}")
self.tts_audio_queue.put((SentenceType.LAST, [], None))
def to_tts(self, text: str) -> list:
@@ -263,7 +265,7 @@ class TTSProvider(TTSProviderBase):
self.api_url, params=params, headers=headers, timeout=5
) as response:
if response.status_code != 200:
logger.error(
logger.bind(tag=TAG).error(
f"TTS请求失败: {response.status_code}, {response.text}"
)
return []
@@ -299,5 +301,5 @@ class TTSProvider(TTSProviderBase):
return opus_datas
except Exception as e:
logger.error(f"TTS请求异常: {e}")
logger.bind(tag=TAG).error(f"TTS请求异常: {e}")
return []
main/xiaozhi-server/core/providers/tts/openai.py
+3 -1
View File
@@ -25,7 +25,9 @@ class TTSProvider(TTSProviderBase):
self.speed = float(speed) if speed else 1.0
self.output_file = config.get("output_dir", "tmp/")
check_model_key("TTS", self.api_key)
model_key_msg = check_model_key("TTS", self.api_key)
if model_key_msg:
logger.bind(tag=TAG).error(model_key_msg)
async def text_to_speak(self, text, output_file):
headers = {
main/xiaozhi-server/core/providers/vllm/openai.py
+3 -1
View File
@@ -34,7 +34,9 @@ class VLLMProvider(VLLMProviderBase):
except (ValueError, TypeError):
setattr(self, param, default)
check_model_key("VLLM", self.api_key)
model_key_msg = check_model_key("VLLM", self.api_key)
if model_key_msg:
logger.bind(tag=TAG).error(model_key_msg)
self.client = openai.OpenAI(api_key=self.api_key, base_url=self.base_url)
def response(self, question, base64_image):
main/xiaozhi-server/core/utils/util.py
+7 -7
View File
@@ -186,10 +186,8 @@ def remove_punctuation_and_length(text):
def check_model_key(modelType, modelKey):
if "你" in modelKey:
raise ValueError(
"你还没配置" + modelType + "的密钥，请检查一下所使用的LLM是否配置了密钥"
)
return True
return f"配置错误: {modelType} 的 API key 未设置,当前值为: {modelKey}"
return None
def parse_string_to_list(value, separator=";"):
@@ -785,7 +783,9 @@ def audio_bytes_to_data(audio_bytes, file_type, is_opus=True):
return p3.decode_opus_from_bytes(audio_bytes)
else:
# 其他格式用pydub
audio = AudioSegment.from_file(BytesIO(audio_bytes), format=file_type, parameters=["-nostdin"])
audio = AudioSegment.from_file(
BytesIO(audio_bytes), format=file_type, parameters=["-nostdin"]
)
audio = audio.set_channels(1).set_frame_rate(16000).set_sample_width(2)
duration = len(audio) / 1000.0
raw_data = audio.raw_data
@@ -838,11 +838,11 @@ def opus_datas_to_wav_bytes(opus_datas, sample_rate=16000, channels=1):
pcm = decoder.decode(opus_frame, frame_size)
pcm_datas.append(pcm)
pcm_bytes = b''.join(pcm_datas)
pcm_bytes = b"".join(pcm_datas)
# 写入wav字节流
wav_buffer = BytesIO()
with wave.open(wav_buffer, 'wb') as wf:
with wave.open(wav_buffer, "wb") as wf:
wf.setnchannels(channels)
wf.setsampwidth(2) # 16bit
wf.setframerate(sample_rate)