Skip to content

Andy963/telegram_ai_bot

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Telegram AI Bot:

This repo a Telegram AI Bot(with chatgpt,azure-openai, palm2 etc.). And it works great.

You can deploy your own bot

Features

  • List of allowed Telegram users
  • use chatgpt-3.5-turbo default
  • reply with voice message (use azure text to speech)
  • multi language voice message support (azure TTS)
  • picture ocr and translate, summary, tell story joke etc.
  • parse a url link(only summary the text for now)
  • paLM2 support(use azure translate to en and ask palm2, palm2 translate is so weak)
  • azure openai support
  • Anthropic claude support
  • user management
  • set stream response by user(for low speed response)

Bot commands

  • /retry – Regenerate last bot answer
  • /new – Start new dialog
  • /mode – Select chat mode
  • /help – Show help
  • /model - List all ai model (chatgpt, palm2, azure_openai etc. admin required)
  • /stream - enable stream response or disable

Setup

  1. Get your OpenAI API , paLM2 key, , azure openai key or claude key.

  2. Get your Telegram bot token from @BotFather

  3. Get your azure free trial account and get your azure key(text,speech,computer vision)

  4. Docker

for docker user:

docker run -d  --name chatgpt -v /etc/gpt:/etc/gpt andy963/telegram_ai_bot:latest

TODO:

  • support voice message. (use azure text to speech)
  • reply with multiple language voice (contains En & Zh, need azure)
  • picture ocr and translate, summary, tell story joke etc.
  • prompt list
  • get content from url and summary the content(weak)
  • paLM2 support
  • export history(only text for now).
  • multi model support (support chatgpt, palm2, azure_openai etc.)
  • azure openai support
  • Anthropic claude support
  • user management
  • set stream response (claude only for now)
  • user list page by page
  • model select by user
  • get the latest new and summarize and send to user on schedule (maybe with voice)

if you are using ubuntu 22.04, you need to install the latest libssl1.1 either as a binary package, or by compiling it from sources.

## References
1. [*Build ChatGPT from GPT-3*](https://learnprompting.org/docs/applied_prompting/build_chatgpt)
2. [install libssl1.1](https://learn.microsoft.com/en-us/azure/cognitive-services/speech-service/quickstarts/setup-platform?pivots=programming-language-python&tabs=linux%2Cubuntu%2Cdotnet%2Cjre%2Cmaven%2Cnodejs%2Cmac%2Cpypi)

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Python 99.5%
  • Dockerfile 0.5%