voice-transcribe

Overview

Transcribe audio files using OpenAI’s gpt-4o-mini-transcribe model with vocabulary hints and text replacements. This skill is useful for developers who receive voice memos and want to automate transcription.

Key Features

• Transcribe audio files using OpenAI’s gpt-4o-mini-transcribe model
• Supports multiple audio formats (mp3, mp4, mpeg, mpga, m4a, wav, webm, ogg, opus)
• Custom vocabulary and text replacements for improved accuracy

How It Works

1. Run the transcribe command with the audio file path as an argument
2. The skill uses OpenAI’s gpt-4o-mini-transcribe model to transcribe the audio file
3. The transcribed text is cached by SHA256 of the audio file for future use

Use Cases

• Transcribe voice memos received via WhatsApp
• Automate transcription for podcast or video recordings
• Improve accuracy by adding custom vocabulary and text replacements