openai-transcribe

name	openai-transcribe
description	Transcribe local audio files with the OpenAI Audio Transcriptions API without adding OpenAI dependencies to the target project.
allowed-tools	Bash(python *)
argument-hint	["audio-file ..."]

OpenAI Transcribe

Use scripts/transcribe.py to transcribe local audio through the OpenAI Audio Transcriptions API.

The script intentionally uses only the Python standard library. It does not add openai or any other dependency to the repository being worked on.

Authentication

The script reads the API key in this order:

1. OPENAI_API_KEY from the current environment.
2. --env-file /path/to/.env, reading OPENAI_API_KEY=....
3. --auto-env, scanning local project .env files under /home/alexey/git and using the first file that contains OPENAI_API_KEY.

The key is never printed.

Example

python /home/alexey/git/.agents/skills/openai-transcribe/scripts/transcribe.py \
  --auto-env \
  --model whisper-1 \
  --language ru \
  --out-dir docs/source-recordings \
  docs/source-recordings/example.m4a

For whisper-1, the script requests verbose_json and writes both the raw JSON response and a timestamped .txt file.

Files larger than the OpenAI upload limit are split with ffmpeg before transcription, then combined back into one timestamped output.