OpenAI Whisper API Transcription

API-based speech-to-text transcription through OpenAI. No local model downloads, no GPU, no Python ML stack — just an API key and a shell script.

Prerequisites

OpenAI Audio API, curl or OpenAI SDK

Installation

Requirements and caveats from upstream:

• Node reference
• python
• speaker labels but requires chunking_strategy when your audio is longer than

Basic usage or getting-started notes:

• Example workflows and tasks teams hand to Codex

• Models

• Pricing

• Source: https://developers.openai.com/api/docs/guides/speech-to-text

Documentation

• https://developers.openai.com/api/docs/guides/speech-to-text

Source

• Agent Skill Exchange