Skip to content
Get packs

OpenAI Whisper Transcription

Transcribe audio files to text using OpenAI's Whisper API via a curl script

Source Download skill
Repository
spinabot/brigade
Stars
11,367
Licence
MIT
Files
2 files

What it does

Once installed, the agent can turn an audio recording (such as m4a or ogg) into a text transcript by sending the file to OpenAI's speech-to-text service. It saves the result as a .txt file by default, or as JSON if asked, and can accept hints like the language or speaker names to improve accuracy. It needs your OpenAI API key, and the audio is uploaded to OpenAI (or to a gateway you configure) for processing.

Great for

  • People who need meeting, interview, or voice-memo recordings turned into text quickly
  • Developers who want a simple shell-based transcription step in an agent workflow
  • Users who already have an OpenAI API account and want to use a self-hosted or OpenAI-compatible proxy

Skip it if

  • Confidential or sensitive recordings, because the audio is sent to a third-party API (OpenAI or whichever gateway OPENAI_BASE_URL points to)
  • Users without an OpenAI API key or who do not want to pay per-use API charges
  • Offline or air-gapped environments, since the skill requires network access to the transcription endpoint

Try asking

  • “Transcribe /Users/me/Downloads/interview.m4a into a text file”
  • “Transcribe this voice memo in English and save the result to /tmp/notes.txt”
  • “Transcribe meeting.ogg with the prompt 'Speaker names: Peter, Daniel' to improve name spelling”
  • “Transcribe podcast.m4a as JSON output with timestamps and metadata”
Step-by-step guide: how to use OpenAI Whisper TranscriptionInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 2 original files, the MIT licence and an attribution note.

    openai-whisper-api.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip openai-whisper-api.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip openai-whisper-api.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Set --language (for example en) when you know the language, which usually improves accuracy and speed
  • Use --prompt to supply names or technical terms that appear in the recording
  • Review the script at scripts/transcribe.sh before first use and avoid uploading sensitive audio unless you trust the destination endpoint

What's inside

  • scripts/transcribe.sh
  • SKILL.md

Redistributed unmodified from spinabot/brigade under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.