AutoTranscriber

repository: AutoTranscriber Auto-Transcriber is an experimental, openai-powered translation tool, generating .srt file in one call!

Main Features

• powered by Openai GPT3.5 or GPT4 model, excellent context and results.
• Audio to Text using Fast-Whisper(a faster implementation of Openai Whisper model), up to 4x faster.
• Minimum ussage of Pandas, using Polars for data processing and generating GPT requests.
• Ascychronize ChatCompletionCall, minimum latency.
• Build with Streamlit, with support to modify transcription results.

How to start with?

git clone //repo
pip install requirements.txt
streamlit run app.py

• Choose From File or From Link from sidebar
• Input yt-link(or video/audio file path), saved folder path
• Click Generate