https://github.com/CoHuK/gpt-telegram-bot
GPT/Whisper/DALL-E Telegram Bot with easy deployment using Chalice to AWS Lambda
https://github.com/CoHuK/gpt-telegram-bot
aws-lambda dall-e gpt gpt-35-turbo telegram telegram-bot whisper
Last synced: over 1 year ago
JSON representation
GPT/Whisper/DALL-E Telegram Bot with easy deployment using Chalice to AWS Lambda
- Host: GitHub
- URL: https://github.com/CoHuK/gpt-telegram-bot
- Owner: CoHuK
- License: other
- Created: 2023-03-21T23:47:44.000Z (over 3 years ago)
- Default Branch: main
- Last Pushed: 2023-11-26T15:36:15.000Z (over 2 years ago)
- Last Synced: 2024-10-24T10:08:05.433Z (over 1 year ago)
- Topics: aws-lambda, dall-e, gpt, gpt-35-turbo, telegram, telegram-bot, whisper
- Language: Python
- Homepage:
- Size: 54 MB
- Stars: 14
- Watchers: 1
- Forks: 0
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- Funding: .github/FUNDING.yml
- License: LICENSE
Awesome Lists containing this project
README
# GPT/Whisper/DALL-E Telegram Bot with easy deployment using Chalice to AWS Lambda
GPT/Whisper/DALL-E Telegram Bot using AWS features
- Lambda, DynamoDB (full auto-deployment)
- ~~Fargate, ECS (manual Docker container deployment)~~ Deprecated
## Requirements
- Python 3.9+
- AWS CLI
- AWS credentials and region setup in the \`~/.aws/credentials\` file
- Chalice
## Deployment to AWS Lambda
1. Create a Free Tier AWS account
2. Install and setup AWS CLI (region must be specified in ~/.aws/credentials. Ex.: region=eu-west-1)
3. Update the \`.chalice/config.json\` file with your Telegram Bot API, OpenAI API Token, and your telegram userID (get it from @userinfobot).
4. Run the `deploy.sh` script to deploy the Telegram Bot to AWS Lambda (or `deploy.sh name` for a custom config)
For voice processing you will need to do the next manually (it will be automated in the next releases):
1. Upload ffmpeg.zip to your S3 bucket
2. Create Lambda Layer using S3 URI
3. Attach layer to your Lambda function
## Features
- Currently used openAI model: gpt-3.5, gpt-4 (Choose in config or by the bot's command)
- Image generation `/image prompt` or just ask without the command!
- Voice message transcript (Just send any voice message)
- Video transcript and translation! (Send any video to the bot, and it will transcribe and translate it to English)
- The context of your chat is saved until you use the command `/clear`
- The price in USD of the response is shown
- Multi-config support: keep multiple chalice configs to deploy multiple bots `.chalice/name.config.json`