cleya transcribe

Transcribe audio to SRT subtitle file (requires Whisper)

Cost: moderate — local Whisper; first run may download a model

cleya transcribe --input video.mp4 --language en

Options

Flag	What it does
--input	Path to input video	required · string
--language	Language code (e.g. en, es, fr, de)	string · default en
--whisper-path	Path to Whisper binary	string · default whisper
--model	Whisper model	string · default base · one of base, small, medium, large
--diarize	Enable speaker diarization (requires tinydiarize model)	boolean
--start	Transcribe from this time (e.g. '30:00')	string
--end	Transcribe until this time (e.g. '40:00')	string
--cloud	Transcribe via the OpenAI cloud API (uploads audio off-device; requires OPENAI_API_KEY). Local-only by default.	boolean
--output	Output SRT file path	string · default <input>.srt
--overwrite	Overwrite output file if it exists	boolean

Category: scout — understand the material before acting

Generated from the toolkit

Generated from cleya capabilities, the toolkit's own manifest, so it cannot drift. Run cleya transcribe --help for the same information in your terminal.