diff --git a/main/xiaozhi-server/config.yaml b/main/xiaozhi-server/config.yaml index f795d8c7..a7c5357b 100644 --- a/main/xiaozhi-server/config.yaml +++ b/main/xiaozhi-server/config.yaml @@ -158,6 +158,15 @@ LLM: base_url: http://homeassistant.local:8123 agent_id: conversation.chatgpt api_key: 你的home assistant api访问令牌 + FastgptLLM: + # 定义LLM API类型 + type: fastgpt + # 如果使用fastgpt，配置文件里prompt(提示词)是无效的，需要在fastgpt控制台设置提示词 + base_url: https://host/api/v1 + api_key: fastgpt-xxx + variables: + k: "v" + k2: "v2" TTS: # 当前支持的type为edge、doubao，可自行适配 EdgeTTS: @@ -316,6 +325,36 @@ TTS: voice: "zh_female_wanwanxiaohe_moon_bigtts" output_file: tmp/ access_token: "你的302API密钥" + ACGNTTS: + #在线网址：https://acgn.ttson.cn/ + #token购买：www.ttson.cn + #开发相关疑问请提交至3497689533@qq.com + #角色id获取地址：ctrl+f快速检索角色——网站管理者不允许发布,可询问网站管理者：1069379506 + #各参数意义见开发文档：https://www.yuque.com/alexuh/skmti9/wm6taqislegb02gd?singleDoc# + type: ttson + token: your_token + voice_id: 1695 + speed_factor: 1 + pitch_factor: 0 + volume_change_dB: 0 + to_lang: ZH + url: https://u95167-bd74-2aef8085.westx.seetacloud.com:8443/flashsummary/tts?token= + format: mp3 + output_file: tmp/ + emotion: 1 + OpenAITTS: + # openai官方文本转语音服务，可支持全球大多数语种 + type: openai + api_key: 你的openai api key + # 国内需要使用代理 + api_url: https://api.openai.com/v1/audio/speech + # 可选tts-1或tts-1-hd，tts-1速度更快tts-1-hd质量更好 + model: tts-1 + # 演讲者，可选alloy, echo, fable, onyx, nova, shimmer + voice: onyx + # 语速范围0.25-4.0 + speed: 1 + output_file: tmp/ # 模块测试配置 module_test: test_sentences: # 自定义测试语句 diff --git a/main/xiaozhi-server/core/providers/llm/fastgpt/fastgpt.py b/main/xiaozhi-server/core/providers/llm/fastgpt/fastgpt.py new file mode 100644 index 00000000..38f9a2c9 --- /dev/null +++ b/main/xiaozhi-server/core/providers/llm/fastgpt/fastgpt.py @@ -0,0 +1,65 @@ +import json +from config.logger import setup_logging +import requests +from core.providers.llm.base import LLMProviderBase + +TAG = __name__ +logger = setup_logging() + + +class LLMProvider(LLMProviderBase): + def __init__(self, config): + self.api_key = config["api_key"] + self.base_url = config.get("base_url") + self.detail = config.get("detail", False) + self.variables = config.get("variables", {}) + + def response(self, session_id, dialogue): + try: + # 取最后一条用户消息 + last_msg = next(m for m in reversed(dialogue) if m["role"] == "user") + + # 发起流式请求 + with requests.post( + f"{self.base_url}/chat/completions", + headers={"Authorization": f"Bearer {self.api_key}"}, + json={ + "stream": True, + "chatId": session_id, + "detail": self.detail, + "variables": self.variables, + "messages": [ + { + "role": "user", + "content": last_msg["content"] + } + ] + }, + stream=True + ) as r: + for line in r.iter_lines(): + if line: + try: + if line.startswith(b'data: '): + if line[6:].decode('utf-8') == '[DONE]': + break + + data = json.loads(line[6:]) + if 'choices' in data and len(data['choices']) > 0: + delta = data['choices'][0].get('delta', {}) + if delta and 'content' in delta and delta['content'] is not None: + content = delta['content'] + if '' in content: + continue + if '' in content: + continue + yield content + + except json.JSONDecodeError as e: + continue + except Exception as e: + continue + + except Exception as e: + logger.bind(tag=TAG).error(f"Error in response generation: {e}") + yield "【服务响应异常】" \ No newline at end of file diff --git a/main/xiaozhi-server/core/providers/llm/gemini/gemini.py b/main/xiaozhi-server/core/providers/llm/gemini/gemini.py index 7ddf84fa..f753ac3e 100644 --- a/main/xiaozhi-server/core/providers/llm/gemini/gemini.py +++ b/main/xiaozhi-server/core/providers/llm/gemini/gemini.py @@ -1,25 +1,24 @@ -from config.logger import setup_logging import google.generativeai as genai +from core.utils.util import check_model_key from core.providers.llm.base import LLMProviderBase -TAG = __name__ -logger = setup_logging() class LLMProvider(LLMProviderBase): def __init__(self, config): """初始化Gemini LLM Provider""" - self.model_name = config.get("model_name", "gemini-1.5-pro") + self.model_name = config.get("model_name", "gemini-1.5-pro") self.api_key = config.get("api_key") - - if not self.api_key or "你" in self.api_key: - logger.bind(tag=TAG).error("你还没配置Gemini LLM的密钥，请在配置文件中配置密钥，否则无法正常工作") + + have_key = check_model_key("LLM", self.api_key) + + if not have_key: return - + try: # 初始化Gemini客户端 genai.configure(api_key=self.api_key) self.model = genai.GenerativeModel(self.model_name) - + # 设置生成参数 self.generation_config = { "temperature": 0.7, @@ -55,7 +54,7 @@ class LLMProvider(LLMProviderBase): # 创建新的聊天会话 chat = self.model.start_chat(history=chat_history) - + # 发送消息并获取流式响应 response = chat.send_message( current_msg, @@ -71,7 +70,7 @@ class LLMProvider(LLMProviderBase): except Exception as e: error_msg = str(e) logger.bind(tag=TAG).error(f"Gemini响应生成错误: {error_msg}") - + # 针对不同错误返回友好提示 if "Rate limit" in error_msg: yield "【Gemini服务请求太频繁,请稍后再试】" diff --git a/main/xiaozhi-server/core/providers/llm/openai/openai.py b/main/xiaozhi-server/core/providers/llm/openai/openai.py index e060248f..55dae20c 100644 --- a/main/xiaozhi-server/core/providers/llm/openai/openai.py +++ b/main/xiaozhi-server/core/providers/llm/openai/openai.py @@ -1,10 +1,7 @@ -from config.logger import setup_logging import openai +from core.utils.util import check_model_key from core.providers.llm.base import LLMProviderBase -TAG = __name__ -logger = setup_logging() - class LLMProvider(LLMProviderBase): def __init__(self, config): @@ -14,8 +11,7 @@ class LLMProvider(LLMProviderBase): self.base_url = config.get("base_url") else: self.base_url = config.get("url") - if "你" in self.api_key: - logger.bind(tag=TAG).error("你还没配置LLM的密钥，请在配置文件中配置密钥，否则无法正常工作") + check_model_key("LLM", self.api_key) self.client = openai.OpenAI(api_key=self.api_key, base_url=self.base_url) def response(self, session_id, dialogue): diff --git a/main/xiaozhi-server/core/providers/memory/mem0ai/mem0ai.py b/main/xiaozhi-server/core/providers/memory/mem0ai/mem0ai.py index 482c0bc2..a904a02d 100644 --- a/main/xiaozhi-server/core/providers/memory/mem0ai/mem0ai.py +++ b/main/xiaozhi-server/core/providers/memory/mem0ai/mem0ai.py @@ -1,5 +1,6 @@ from ..base import MemoryProviderBase, logger from mem0 import MemoryClient +from core.utils.util import check_model_key TAG = __name__ @@ -8,8 +9,8 @@ class MemoryProvider(MemoryProviderBase): super().__init__(config) self.api_key = config.get("api_key", "") self.api_version = config.get("api_version", "v1.1") - if len(self.api_key) == 0 or "你" in self.api_key: - logger.bind(tag=TAG).error("你还没配置Mem0ai的密钥，请在配置文件中配置密钥，否则无法提供记忆服务") + have_key = check_model_key("Mem0ai", self.api_key) + if not have_key : self.use_mem0 = False return else: diff --git a/main/xiaozhi-server/core/providers/tts/doubao.py b/main/xiaozhi-server/core/providers/tts/doubao.py index 3b812131..55438a0c 100644 --- a/main/xiaozhi-server/core/providers/tts/doubao.py +++ b/main/xiaozhi-server/core/providers/tts/doubao.py @@ -4,6 +4,7 @@ import json import base64 import requests from datetime import datetime +from core.utils.util import check_model_key from core.providers.tts.base import TTSProviderBase @@ -17,6 +18,7 @@ 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) def generate_filename(self, extension=".wav"): return os.path.join(self.output_file, f"tts-{datetime.now().date()}@{uuid.uuid4().hex}{extension}") diff --git a/main/xiaozhi-server/core/providers/tts/fishspeech.py b/main/xiaozhi-server/core/providers/tts/fishspeech.py index 5794a2e3..e8af4eb4 100644 --- a/main/xiaozhi-server/core/providers/tts/fishspeech.py +++ b/main/xiaozhi-server/core/providers/tts/fishspeech.py @@ -1,4 +1,3 @@ - import base64 import os import uuid @@ -9,6 +8,7 @@ from pydantic import BaseModel, Field, conint, model_validator from typing_extensions import Annotated from datetime import datetime from typing import Literal +from core.utils.util import check_model_key from core.providers.tts.base import TTSProviderBase from config.logger import setup_logging @@ -24,7 +24,7 @@ class ServeReferenceAudio(BaseModel): def decode_audio(cls, values): audio = values.get("audio") if ( - isinstance(audio, str) and len(audio) > 255 + isinstance(audio, str) and len(audio) > 255 ): # Check if audio is a string (Base64) try: values["audio"] = base64.b64decode(audio) @@ -36,6 +36,7 @@ class ServeReferenceAudio(BaseModel): def __repr__(self) -> str: return f"ServeReferenceAudio(text={self.text!r}, audio_size={len(self.audio)})" + class ServeTTSRequest(BaseModel): text: str chunk_length: Annotated[int, conint(ge=100, le=300, strict=True)] = 200 @@ -70,6 +71,7 @@ def audio_to_bytes(file_path): wav = wav_file.read() return wav + def read_ref_text(ref_text): path = Path(ref_text) if path.exists() and path.is_file(): @@ -77,31 +79,32 @@ def read_ref_text(ref_text): return file.read() return ref_text + class TTSProvider(TTSProviderBase): def __init__(self, config, delete_audio_file): super().__init__(config, delete_audio_file) self.reference_id = config.get("reference_id") - self.reference_audio = config.get("reference_audio",[]) - self.reference_text = config.get("reference_text",[]) - self.format = config.get("format","wav") - self.channels = config.get("channels",1) - self.rate = config.get("rate",44100) - self.api_key = config.get("api_key","YOUR_API_KEY") - if "你" in self.api_key: - logger.bind(tag=TAG).error("你还没配置FishSpeech TTS的密钥，请在配置文件中配置密钥，否则无法正常工作") + self.reference_audio = config.get("reference_audio", []) + self.reference_text = config.get("reference_text", []) + self.format = config.get("format", "wav") + self.channels = config.get("channels", 1) + self.rate = config.get("rate", 44100) + self.api_key = config.get("api_key", "YOUR_API_KEY") + have_key = check_model_key("FishSpeech TTS", self.api_key) + if not have_key: return - self.normalize = config.get("normalize",True) - self.max_new_tokens = config.get("max_new_tokens",1024) - self.chunk_length = config.get("chunk_length",200) - self.top_p = config.get("top_p",0.7) - self.repetition_penalty = config.get("repetition_penalty",1.2) - self.temperature = config.get("temperature",0.7) - self.streaming = config.get("streaming",False) - self.use_memory_cache = config.get("use_memory_cache","on") + self.normalize = config.get("normalize", True) + self.max_new_tokens = config.get("max_new_tokens", 1024) + self.chunk_length = config.get("chunk_length", 200) + self.top_p = config.get("top_p", 0.7) + self.repetition_penalty = config.get("repetition_penalty", 1.2) + self.temperature = config.get("temperature", 0.7) + self.streaming = config.get("streaming", False) + self.use_memory_cache = config.get("use_memory_cache", "on") self.seed = config.get("seed") - self.api_url = config.get("api_url","http://127.0.0.1:8080/v1/tts") + self.api_url = config.get("api_url", "http://127.0.0.1:8080/v1/tts") def generate_filename(self, extension=".wav"): return os.path.join(self.output_file, f"tts-{datetime.now().date()}@{uuid.uuid4().hex}{extension}") @@ -148,11 +151,9 @@ class TTSProvider(TTSProviderBase): with open(output_file, "wb") as audio_file: audio_file.write(audio_content) - + else: print(f"Request failed with status code {response.status_code}") print(response.json()) - - diff --git a/main/xiaozhi-server/core/providers/tts/openai.py b/main/xiaozhi-server/core/providers/tts/openai.py new file mode 100644 index 00000000..dbdf311f --- /dev/null +++ b/main/xiaozhi-server/core/providers/tts/openai.py @@ -0,0 +1,40 @@ +import os +import uuid +import requests +from datetime import datetime +from core.utils.util import check_model_key +from core.providers.tts.base import TTSProviderBase + +class TTSProvider(TTSProviderBase): + def __init__(self, config, delete_audio_file): + super().__init__(config, delete_audio_file) + self.api_key = config.get("api_key") + self.api_url = config.get("api_url", "https://api.openai.com/v1/audio/speech") + self.model = config.get("model", "tts-1") + self.voice = config.get("voice", "alloy") + self.response_format = "wav" + self.speed = config.get("speed", 1.0) + self.output_file = config.get("output_file", "tmp/") + check_model_key("TTS", self.api_key) + + def generate_filename(self, extension=".wav"): + return os.path.join(self.output_file, f"tts-{datetime.now().date()}@{uuid.uuid4().hex}{extension}") + + async def text_to_speak(self, text, output_file): + headers = { + "Authorization": f"Bearer {self.api_key}", + "Content-Type": "application/json" + } + data = { + "model": self.model, + "input": text, + "voice": self.voice, + "response_format": "wav", + "speed": self.speed + } + response = requests.post(self.api_url, json=data, headers=headers) + if response.status_code == 200: + with open(output_file, "wb") as audio_file: + audio_file.write(response.content) + else: + raise Exception(f"OpenAI TTS请求失败: {response.status_code} - {response.text}") diff --git a/main/xiaozhi-server/core/providers/tts/ttson.py b/main/xiaozhi-server/core/providers/tts/ttson.py new file mode 100644 index 00000000..e56fd02d --- /dev/null +++ b/main/xiaozhi-server/core/providers/tts/ttson.py @@ -0,0 +1,64 @@ +import os +import uuid +import json +import requests +import shutil +from datetime import datetime +from core.providers.tts.base import TTSProviderBase + + +class TTSProvider(TTSProviderBase): + def __init__(self, config, delete_audio_file): + super().__init__(config, delete_audio_file) + self.url = config.get("url", "https://u95167-bd74-2aef8085.westx.seetacloud.com:8443/flashsummary/tts?token=") + self.voice_id = config.get("voice_id", 1695) + self.token = config.get("token") + self.to_lang = config.get("to_lang") + self.volume_change_dB = config.get("volume_change_dB", 0) + self.speed_factor = config.get("speed_factor", 1) + self.stream = config.get("stream", False) + self.output_file = config.get("output_file") + self.pitch_factor = config.get("pitch_factor", 0) + self.format = config.get("format", "mp3") + self.emotion = config.get("emotion", 1) + self.header = { + "Content-Type": "application/json" + } + + def generate_filename(self, extension=".mp3"): + return os.path.join(self.output_file, f"tts-{datetime.now().date()}@{uuid.uuid4().hex}{extension}") + + async def text_to_speak(self, text, output_file): + url = f'{self.url}{self.token}' + result = "firefly" + payload = json.dumps({ + "to_lang": self.to_lang, + "text": text, + "emotion": self.emotion, + "format": self.format, + "volume_change_dB": self.volume_change_dB, + "voice_id": self.voice_id, + "pitch_factor": self.pitch_factor, + "speed_factor": self.speed_factor, + "token": self.token + }) + + resp = requests.request("POST", url, data=payload) + if resp.status_code != 200: + return None + resp_json = resp.json() + try: + result = resp_json['url'] + ':' + str( + resp_json[ + 'port']) + '/flashsummary/retrieveFileData?stream=True&token=' + self.token + '&voice_audio_path=' + \ + resp_json['voice_path'] + except Exception as e: + print("error:", e) + + audio_content = requests.get(result) + with open(output_file, "wb") as f: + f.write(audio_content.content) + return True + voice_path = resp_json.get("voice_path") + des_path = output_file + shutil.move(voice_path, des_path) diff --git a/main/xiaozhi-server/core/utils/util.py b/main/xiaozhi-server/core/utils/util.py index df792cbe..4bc37d28 100644 --- a/main/xiaozhi-server/core/utils/util.py +++ b/main/xiaozhi-server/core/utils/util.py @@ -1,9 +1,9 @@ import os -import re import json import yaml import socket import subprocess +import logging def get_project_dir(): @@ -87,35 +87,13 @@ def remove_punctuation_and_length(text): return 0, "" return len(result), result - -def check_password(password): - """ - 检查密码是否满足以下条件： - 1. 密码长度大于八位。 - 2. 密码包含英文和数字。 - 3. 密码不能包含“xiaozhi”字符。 - - :param password: 要检查的密码 - :return: 如果密码满足条件，则返回True；否则返回False。 - """ - # 检查密码长度 - if len(password) < 8: +def check_model_key(modelType, modelKey): + if "你" in modelKey: + logging.error("你还没配置" + modelType + "的密钥，请在配置文件中配置密钥，否则无法正常工作") return False - - # 检查是否包含英文字符和数字 - if not re.search(r'[A-Za-z]', password) or not re.search(r'[0-9]', password): - return False - - # 检查是否包含“xiaozhi”字符 - if "xiaozhi" in password: - return False - - if "1234" in password: - return False - - # 如果满足所有条件，则返回True return True + def check_ffmpeg_installed(): ffmpeg_installed = False try: @@ -140,4 +118,4 @@ def check_ffmpeg_installed(): error_msg += "\n建议您：\n" error_msg += "1、按照项目的安装文档，正确进入conda环境\n" error_msg += "2、查阅安装文档，如何在conda环境中安装ffmpeg\n" - raise ValueError(error_msg) \ No newline at end of file + raise ValueError(error_msg)