Whisper

Overview

Transcribe audio with OpenAI's Whisper — the state-of-the-art speech recognition model. This skill covers local Whisper (Python), faster-whisper (CTranslate2, 4x faster), whisper.cpp (CPU-optimized C++), and the OpenAI Whisper API. Includes subtitle generation (SRT/VTT/JSON), multi-language transcription, translation to English, speaker diarization, word-level timestamps, and production pipeline patterns for podcasts, meetings, and video subtitles.

Instructions

Step 1: Choose Your Runtime

Option A — OpenAI Whisper (original Python):

pip install openai-whisper
# Models: tiny (39M), base (74M), small (244M), medium (769M), large-v3 (1.5G)

Option B — faster-whisper (recommended for local, 4x faster):

pip install faster-whisper