search

FoloToy / folotoy-server-self-hosting

Config files for self-hosting the FoloToy Server. Documents: https://docs.folotoy.com
https://tool.folotoy.com
GNU General Public License v3.0
415 stars 75 forks source link

OPENAI_WHISPER_API_BASE没有生效 #121

Closed aidots closed 4 months ago

aidots commented 4 months ago

folotoy-server-self-hosting-folotoy-1 | 2024-02-27 20:41:13,101 - ThreadPoolExecutor-0_2 - util.py(60) - DEBUG - message='Request to OpenAI API' method=post path=https://api.openai.com/v1/audio/transcriptions folotoy-server-self-hosting-folotoy-1 | 2024-02-27 20:41:13,101 - ThreadPoolExecutor-0_2 - util.py(60) - DEBUG - api_version=None data='{\'model\': \'whisper-1\', \'prompt\': "ok.let\'s begin."}' message='Post details'

aidots commented 4 months ago

配置：

version: '3'
volumes:
  emqx-etc:
  emqx-data:
  emqx-log:
services:
  emqx:
    image: emqx/emqx:latest
    restart: always
    ports:
      - "1883:1883/tcp"
      - "18083:18083/tcp"
      - "8083:8083/tcp"
    volumes:
      - emqx-etc:/opt/emqx/etc    
      - emqx-data:/opt/emqx/data
      - emqx-log:/opt/emqx/log
    environment:
      EMQX_NODE_NAME: emqx@node1.emqx.io
  nginx:
    image: nginx:latest
    restart: always
    ports:
      - "8082:80/tcp"
    volumes:
      - ./audio:/usr/share/nginx/html
  folotoy:
    image: lewangdev/folotoy-server:latest
    restart: always
    ports:
      - "8085:8085/udp"
    volumes:
      - ./audio:/audio
      - ./config:/config
    environment:
      TZ: Asia/Shanghai

      LOG_LEVEL: DEBUG

      ROLES_FILE_PATH: /config/roles.json

      # Default STT(Sound To Text) type
      # Options: [openai-whisper, azure-whisper, azure-stt, dify-stt, aliyun-asr]
      STT_TYPE: openai-whisper

      # OpenAI Whisper
      OPENAI_WHISPER_API_BASE: https://xxx.xxxxxxx.com/v1
      OPENAI_WHISPER_KEY: sk-xxxxxxxxxxxxxxxxx
      OPENAI_WHISPER_MODEL: whisper-1

      # Azure Whisper
      AZURE_WHISPER_API_BASE: https://xxxxx.openai.azure.com
      AZURE_WHISPER_KEY: 9afbef65bcf6487eeeeeeeeeeeeeeeeeee
      AZURE_WHISPER_DEPLOYMENT_NAME: whisper
      AZURE_WHISPER_API_VERSION: 2023-09-01-preview

      # Azure STT
      AZURE_STT_KEY: 3eba91b6143f4d3eeeeeeeeeeeeeeeeeeeeeeeee
      AZURE_STT_SERVICE_REGION: eastasia

      # Dify STT
      DIFY_STT_API_BASE: https://api.dify.ai/v1
      DIFY_STT_KEY: app-xxxxxxxxxxxxxxxxxxxxx

      # Aliyun ASR
      ALIYUN_ASR_ACCESS_KEY_ID: LTxxxxxxxxxxxxxxxxxxxxx
      ALIYUN_ASR_ACCESS_KEY_SECRET: 3zWkHVxxxxxxxxxxxxxxxxxxxxx
      ALIYUN_ASR_APP_KEY: Ltamxxxxxxxxxxxxxxxxxxxxx

      # Default LLM(Large Language Model) type
      # Options: [openai, azure-openai, gemini, dify, qianfan]
      LLM_TYPE: openai

      # OpenAI
      OPENAI_OPENAI_API_BASE: https://api.moonshot.cn/v1
      OPENAI_OPENAI_MODEL: moonshot-v1-8k
      OPENAI_OPENAI_KEY: xxxxxxxxxxxxxxxxxxxxxx

      #Azure OpenAI
      AZURE_OPENAI_KEY: xxxxxxxxxxxxxxxxxxxxxx
      AZURE_OPENAI_ENDPOINT: https://xxx.openai.azure.com/
      AZURE_OPENAI_API_VERSION: "2023-05-15"
      #AZURE_OPENAI_MODEL: gpt-35-turbo

      #DIFY
      DIFY_API_BASE: https://api.dify.ai/v1
      DIFY_KEY: app-5YGbxxxxxxxxxxxxxxxxxxxxx

      #Xiaodu API
      XIAODU_CLIENT_ID: xxxxxxxxxxxxxxxxxx
      XIAODU_SECRET: xxxxxxxxxxxxxxxxxxxxx

      # GEMINI
      GEMINI_KEY: AIzaSyxxxxxxxxxxxxxxxxxxk

      # Qianfan
      QIANFAN_CLIENT_ID: 13rBTgxxxxxxxxxxxxxxxxxx
      QIANFAN_SECRET: zYxtMIQLexxxxxxxxxxxxxxxxxx

      # If your elevenlabs is a free account, keep 2 here
      VOICE_EXECUTOR_MAX_WORKERS: 2

      # Default TTS(Text to Sound) type
      # Options: [openai-tts, azure-tts, elevenlabs, edge-tts, aliyun-tts]
      # edge-tts is Free but slow
      TTS_TYPE: azure-tts

      # Azure TTS
      AZURE_TTS_KEY: xxxxxxxxxxxxxxxxxxxxxxxx
      AZURE_TTS_SERVICE_REGION: southeastasia

      # elevenlabs
      ELEVENLABS_TTS_KEY: a920b73991e68d5c9c9aaaaaaaaaaaaaaaa
      ELEVENLABS_TTS_MODEL: eleven_multilingual_v2

      # OpenAI TTS
      OPENAI_TTS_KEY: sk-16XnP3HLHWho21oO2m0AAAAAAAAAAAAAAAAAAAAAA
      OPENAI_TTS_MODEL: tts-1  

      # Aliyun TTS
      ALIYUN_TTS_ACCESS_KEY_ID: LTAI5t91NSxxxxxxxxxxxxxxxxxxxxx
      ALIYUN_TTS_ACCESS_KEY_SECRET: 3zWkHVxxxxxxxxxxxxxxxxxxxxx
      ALIYUN_TTS_APP_KEY: Ltamxxxxxxxxxxxxx

      AUDIO_DOWNLOAD_URL: http://xxxxxxxxxxxx:8082
      AUDIO_SAVE_PATH: /audio

      # MQTT Broker
      MQTT_BROKER_HOST: emqx
      MQTT_BROKER_PORT: 1883
      MQTT_CLIENT_ID: folotoy
      MQTT_USERNAME: folotoy
      MQTT_PASSWORD: folotoy

      SPEECH_UDP_SERVER_HOST: xxxxxxxxxxxxxxx
      SPEECH_UDP_SERVER_PORT: 8085
lewangdev commented 4 months ago

贴一下 roles.json

aidots commented 4 months ago 
{
  "1": {
    "model": "gpt-3.5-turbo",
    "start_text": "你好，我是小兔兔，请问有什么我可以帮助你的吗？",
    "prompt": "你扮演一个孩子的小伙伴，名字叫小兔兔，性格和善，说话活泼可爱，对孩子充满爱心，经常赞赏和鼓励孩子，用5岁孩子容易理解语言提供有趣和创新的回答，每次回复根据聊天主题询问她的看法以激发她的思考和好奇心，现在她来到了你身边问了第一个问题:[你是谁]",
    "max_message_count": 20,
    "temperature": 0.7,
    "max_tokens": 800,
    "top_p": 0.95,
    "frequency_penalty": 0,
    "presence_penalty": 0,
    "voice_name": "zh-CN-XiaoxiaoNeural",
    "language": "zh"
  },
  "2": {
    "model": "gpt-3.5-turbo",
    "start_text": "你好，俺是东北兔，请问有什么俺可以帮助你的吗？",
    "prompt": "你是一个知识渊博，乐于助人的智能机器人,你的名字叫“东北兔”，你的任务是陪我聊天，请用简短的对话方式，用中文讲一段话，每次回答不超过50个字！",
    "max_message_count": 20,
    "temperature": 0.7,
    "max_tokens": 800,
    "top_p": 0.95,
    "frequency_penalty": 0,
    "presence_penalty": 0,
    "voice_name": "zh-CN-liaoning-XiaobeiNeural",
    "language": "zh"
  },
  "3": {
    "model": "gpt-3.5-turbo",
    "start_text": "Hi, I'm Fofo. Nice to meet you.",
    "prompt": "Your name is \"Fofo\". Your task is to chat with me. Please respond in English, keeping your answers brief – no more than 50 words each time!",
    "max_message_count": 20,
    "temperature": 0.7,
    "max_tokens": 800,
    "top_p": 0.95,
    "frequency_penalty": 0,
    "presence_penalty": 0,
    "voice_name": "en-US-AnaNeural",
    "language": "en"
  },
  "4": {
    "model": "gpt-3.5-turbo",
    "start_text": "你好，额是陕西兔，请问有什么额可以帮助你的吗？ ",
    "prompt": "你擅于鼓励别人，乐观积极，无论别人和你说了什么，你都能夸对方，让人快乐",
    "max_message_count": 10,
    "temperature": 0.7,
    "max_tokens": 800,
    "top_p": 0.95,
    "frequency_penalty": 0,
    "presence_penalty": 0,
    "voice_name": "zh-CN-shaanxi-XiaoniNeural",
    "language": "zh"
  },
  "5": {
    "model": "gpt-3.5-turbo",
    "start_text": "你好，我是童话故事兔，想听什么童话故事吗？试试说，我想听听三只小猫咪的故事",
    "prompt": "你是一个知识渊博的智能机器人,你的名字叫“故事兔”，你的任务是讲故事给一位7岁孩子的听，你要先听取孩子想听的故事主题，然后根据孩子的说的内容，用中文讲一段故事，每个故事不超过200个字！",
    "max_message_count": 10,
    "temperature": 0.7,
    "max_tokens": 800,
    "top_p": 0.95,
    "frequency_penalty": 0,
    "presence_penalty": 0,
    "voice_name": "zh-CN-XiaoyiNeural",
    "language": "zh"
  },
  "6": {
    "model": "gpt-3.5-turbo",
    "start_text": "你好，我是台湾兔，请问有什么我可以帮助你的吗？",
    "prompt": "你是一个知识渊博，乐于助人的智能机器人,你的名字叫“台湾兔”，你的任务是陪我聊天，请用简短的对话方式，用中文讲一段话，每次回答不超过50个字！",
    "max_message_count": 10,
    "temperature": 0.7,
    "max_tokens": 800,
    "top_p": 0.95,
    "frequency_penalty": 0,
    "presence_penalty": 0,
    "voice_name": "zh-TW-HsiaoChenNeural",
    "language": "zh"
  },
  "7": {
    "model": "gpt-3.5-turbo",
    "start_text": "你好，我是口算兔，我们一起来玩玩口算游戏吧？",
    "prompt": "我是一个6岁小朋友，你陪我玩口算游戏。你出题，我回答结果。如果答对了你就说好棒，答错了你就告诉我正确答案，并且鼓励我。你一题一题的出，我一个个回答。不要有太多的解释说明。明白了吗？",
    "max_message_count": 20,
    "temperature": 0.7,
    "max_tokens": 800,
    "top_p": 0.95,
    "frequency_penalty": 0,
    "presence_penalty": 0,
    "voice_name": "zh-CN-YunxiaNeural",
    "language": "zh"
  }
}
lewangdev commented 4 months ago

docker-compose.yml 所在的目录下执行 docker compose -d

lewangdev commented 4 months ago

用什么方式部署的？1panel 还是命令行？

aidots commented 4 months ago

命令行

aidots commented 4 months ago

docker-compose.yml 所在的目录下执行 docker compose -d

已经执行过 sudo docker compose up -d

lewangdev commented 4 months ago

稍等，我测试一下

lewangdev commented 4 months ago

我们用你的配置测试没有问题，你的服务器方便加我的 ssh 公钥吗：https://github.com/lewangdev.keys

然后把登录ip/port 发到我的邮箱，我帮你看一下 lewang.dev@gmail.com

aidots commented 4 months ago

运行了这个后才可以： docker compose pull folotoy docker compose up folotoy -d