An Example of Using OpenAI Whisper Transcription API

Batch transcribes audio files in the specified folder.

Prerequisites

• Python 3
• OpenAI API key of an active account.

Setup

For MacOS or Linux:

# setup python environment
python3 -m venv venv
source venv/bin/activate
pip install -U pip
pip install -r requirements.txt

# set api key as environment variable
echo "export OPENAI_API_KEY=<your-openai-api-key>" \
 >> ~/.zshrc # use ~/.bashrc if bash
source ~/.zshrc

Usage

The folder containing the audio files is the 'source folder', the folder that will store the transcription results as txt files is the 'export folder'.

python script.py <source-folder-path> <export-folder-path>

Example:

python script.py ~/Downloads/records/ transcripts/

Support

Available audio file types:

• flac
• mp3
• mp4
• mpeg
• mpga
• m4a
• ogg
• wav
• webm