transcription

name	transcription
description	Audio and video transcription using Whisper AI via the transcribe-anything package. Supports local files, YouTube URLs, and microphone input with multiple backends (faster-whisper, openai-whisper, Whisper API).
version	1.1.0
model	sonnet
invoked_by	both
user_invocable	true
tools	["Bash","Read","Write"]
agents	["developer","ai-ml-expert","researcher"]
lastVerifiedAt	"2026-03-15T00:00:00.000Z"
category	AI/ML
tags	["transcription","audio","video","whisper","speech-to-text","faster-whisper","subtitles","srt"]
best_practices	["Use faster-whisper backend by default — it is 4x faster than openai-whisper on the same hardware","Select model size based on quality/speed tradeoff (tiny→large-v3)","Always specify output directory to keep files organized","Use --lang to skip auto-detection when the source language is known"]
error_handling	strict
source	builtin
trust_score	100
provenance_sha	null

Transcription

Overview

Transcribe audio/video files (local or remote) using Whisper AI via transcribe-anything. Supports local files, YouTube URLs, and microphone input. Output formats: SRT, VTT, plain text, JSON.

Installation

pip install transcribe-anything

Backends install automatically in isolated virtual environments.

Usage

# Local file
transcribe-anything audio.mp3

# YouTube URL
transcribe-anything "https://www.youtube.com/watch?v=VIDEO_ID"

# With options
transcribe-anything audio.mp3 --model large-v3 --lang en --output_dir ./transcripts/

# GPU / device selection
transcribe-anything audio.mp3 --device cuda      # NVIDIA GPU
transcribe-anything audio.mp3 --device mlx       # Mac Apple Silicon (fastest on Mac)
transcribe-anything audio.mp3 --device groq      # Cloud API (fastest overall)

# Speaker diarization (requires HuggingFace token)
transcribe-anything audio.mp3 --device insane --hf_token YOUR_HF_TOKEN

Key Options

Option	Description	Default
--model	tiny, small, medium, large, large-v3	large-v3
--lang	Language code (en, fr, de) or auto	auto-detect
--device	cpu, cuda, mlx, insane, groq	auto-select
--output_dir	Directory to write transcript files	./
--task	transcribe or translate (→ English)