gpt-3.5-turbo

Usage of whisper-1 and gpt-3.5-turbo

This is a sample to showcase how audio can be transcribed using whisper-1 model followed by invoking gpt-3.5-turbo to respond to a query. It demonstrates prompt engineering and basic usage of OpenAI models

Pre-reqs

• To capture audio, ensure you have ffmpeg installed.

  1. Mac

  brew install ffmpeg

  2. Windows

• python3 with pip

Local run

pip install -r requirements.txt

python main.py