Captions4 operations

`speech-captions`

Transcribe speech to timed captions with faster-whisper, detect spoken language, and build SRT/VTT from text or segments. Use when you need automatic captions from audio/video, language detection before transcription, or subtitle files from transcript JSON.

View on GitHub Install skill

Overview

Requires ffmpeg on PATH. Transcription scripts need Python 3.11+ with faster-whisper (installed automatically via `uv run`). CPU-friendly `tiny` model is the default.

Recipes

Copy-paste command sequences agents can run via uv run.

Transcribe a clip to SRT + JSON

uv run scripts/transcribe.py --input interview.mp4 --model tiny

Detect language before choosing a model

uv run scripts/detect-language.py --input interview.mp4

Build SRT from transcript text (no ASR)

uv run scripts/to-srt.py --text "Hello world. This is a test." --duration 6 --output captions.srt

Build VTT from transcribe JSON

uv run scripts/to-vtt.py --segments-json interview_transcript.json --output captions.vtt

End-to-end caption pipeline

Operations

Each script returns structured JSON — see the script contract.

Op ID	Script	Common flags
`speech_captions.detect_language`	`detect-language.py`	--input
`speech_captions.to_srt`	`to-srt.py`	--input
`speech_captions.to_vtt`	`to-vtt.py`	--input
`speech_captions.transcribe`	`transcribe.py`	--input

Do not use for

Burning subtitles into video (use `subtitles/burn.py`)
Broadcast SCC/SMPTE-TT export (use `captions-compliance`)
Video or audio transforms (use `video-transformation` / `audio`)

Related cookbooks

Broadcast captions package