LINE Speech to Text

This is a minimal self-hosted LINE bot using OpenAI's Whisper to transcribe audio messages to text. It runs on Flask and use Gunicorn as WSGI server. Change initial_prompt to fit your need (language, puctuation, names, etc.)

Setup

pip install flask openai-whisper gunicorn

Grab your Channel secret and Channel access token, than replace the placeholders in line_stt.py.

Run

gunicorn wsgi:app