mcpField
G
groq-stt

groq-stt

github.com/958877748Creative & media080 installsUpdated 12w ago
Repository

About groq-stt

groq-stt is a skill that uploads audio to the Groq Speech-to-Text API and saves the resulting transcript beside the input file. It is for users who need a simple file-based transcription workflow, with whisper-large-v3-turbo as the default model.

Key features

  1. 01Transcribe audio through Groq API
  2. 02Save transcript beside source audio
  3. 03Use whisper-large-v3-turbo by default
  4. 04Support common audio and video formats

Use cases

  1. 01Create a text transcript from an MP4 recording
  2. 02Transcribe an audio file from the command line

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/958877748/skills --skill groq-stt

After installing, type/groq-sttin Claude Code to confirm it loaded.

Bundled filesscripts/transcribe.mjs

Details

From the project README

Groq STT Skill

This skill uploads an audio file to the Groq Speech-to-Text API and saves the transcription.

Usage

# set your API key (or use .env in the repo root)
export GROQ_API_KEY=your_api_key_here

# run the script with a path to an audio file
node scripts/transcribe.mjs /path/to/audio.mp4

Output

  • Writes a {filename}_transcript.txt next to the input file.

Notes

  • Uses the whisper-large-v3-turbo model by default.
  • Supported file types: flac, mp3, mp4, mpeg, mpga, m4a, ogg, wav, webm

Comments

Markdown supported