Which audio formats are supported by this skill?

The skill supports common formats compatible with the OpenAI API, such as .m4a, .mp3, .wav, and .ogg.

Do I need an OpenAI API key to use this?

Yes, you must set the OPENAI_API_KEY environment variable or configure it in your clawdis.json file.

Can I specify the transcription language?

Yes, you can use the --language flag to define the source language for better accuracy and fewer errors.

What is the benefit of using a custom prompt?

Providing a prompt helps the model recognize specific technical terms, unusual spellings, or the names of participants in the audio.

Does this skill require any local dependencies?

This skill requires 'curl' to be installed on your system to handle the communication with OpenAI's transcription endpoint.

OpenAI Whisper Transcription

Name: OpenAI Whisper Transcription
Author: steipete

bysteipete

•

432

데이터 과학 및 ML

Transcribes audio files into text or JSON format using OpenAI's state-of-the-art Whisper API.

소개

This skill integrates OpenAI's Whisper model into the Claude Code environment, enabling developers to programmatically convert audio recordings into accurate text transcripts. It supports various file formats like M4A and OGG, allows for custom prompts to improve accuracy—such as specifying speaker names—and offers flexible output options including plain text or structured JSON. It is particularly useful for processing voice notes, transcribing meeting recordings, or building automated voice-to-text pipelines within the Clawdis assistant framework.

주요 기능

Custom prompt injection for better terminology and speaker identification
High-accuracy transcription using the OpenAI Whisper-1 model
Support for multiple audio formats including M4A, OGG, and MP3
Multi-language support with explicit language tagging for improved results
Flexible output formats including plain text and machine-readable JSON
432 GitHub stars

사용 사례

Automating the transcription of voice messages from messaging platforms
Converting meeting recordings into searchable documentation
Extracting structured metadata from audio for data science projects

소개

주요 기능

Custom prompt injection for better terminology and speaker identification
High-accuracy transcription using the OpenAI Whisper-1 model
Support for multiple audio formats including M4A, OGG, and MP3
Multi-language support with explicit language tagging for improved results
Flexible output formats including plain text and machine-readable JSON
432 GitHub stars

사용 사례

Automating the transcription of voice messages from messaging platforms
Converting meeting recordings into searchable documentation
Extracting structured metadata from audio for data science projects