/ai-chatbot

Integrate ChatGPT, Dall-E, Whisper and other AI models in Replicate into Messenger and Telegram bot

Primary LanguageTypeScript

Messenger + Telegram bot with AI

Installation

mkdir downloads

# Install ffmpeg package
# Windows
choco install ffmpeg
# Ubuntu
sudo apt install ffmpeg

cp .env.example .env
# Edit .env

yarn

yarn run start

Telegram commands

ai - My assistant
apps - Show all apps
chat - ChatGPT bot
new - New conversation
new_char - New conversation with current character
imagine - Create image using Dall-E 2
speak - Text to speech
save - Save conversation
settings - Settings
debug - Debug
help - How to use this bot?

Changelog

24/3

  • Add Telegram bot with ChatGPT

28/3

  • Add voice chat to Telegram bot

8/4

  • Update Apps + Settings - easier to config

Configuration

The bottender.config.js File

Bottender configuration file. You can use this file to provide settings for the session store and channels.

The .env File

Bottender utilizes the dotenv package to load your environment variables when developing your app.

To make the bot work, you must put required environment variables into your .env file.

Available Scripts

In the project directory, you can run:

npm run dev

Runs the app in development mode.
The bot will automatically reload if you make changes to the code.
By default, server runs on http://localhost:5000 and ngrok runs on http://localhost:4040.

To run in Console Mode, provide the --console option:

npm run dev -- --console
yarn dev --console

npm start

Runs the app in production mode.
By default, server runs on http://localhost:5000.

To run in Console Mode, provide the --console option:

npm start -- --console
yarn start --console

npm run lint

Runs the linter rules using Eslint.

npm test

Runs the test cases using Jest.

Learn More

To learn Bottender, check out the Bottender documentation.

For more examples, see Bottender examples.