silvanojr / WhatsAppTranscription

Audio Transcription for WhatsApp

Update .env with your OpenAI API key and make sure you are using NodeJS >= 16.

You can find more information about OpenAI here: https://platform.openai.com/overview

You need ffmpeg installed to convert audio files: https://ffmpeg.org

npm install

node index.js

The console will show the QR code that you need to read using WhatsApp's option to link a device. When you link your device, you need to wait for the application to sync with WhatsApp. This may take a while. It will be ready when you see this message in the console:

🚀 @OPEN-WA ready for account: XXXX

Enjoy.

Feature

• It's possible to manually force an audio to be read using the variable SECRET_WORD. Default is "!ler"

Docker

docker build . -t whats

Docker Compose

version: "3.5"
services:
  speech2text:
    image: whats
    environment:
       - OPENAI_API_KEY=sk-xxxxxxx
       - GROUPS=[e-mail],[e-mail]
       - PATH_MP3=/mp3
       - PATH_SESSION=/session
       - SECRET_WORD=!ler
    volumes:
       - session:/session
       - mp3:/mp3
volumes:
  session:
  mp3: