caption kitchenTutorial audio transcription
Transcribe a short audio clip into text or WebVTT using cloud inference.
Audio to captions
Transcribe a short audio clip into text or WebVTT using cloud inference.
A little possibility.
Add your details on the left. Your next starting point will appear here.
01Bring your details
A few useful facts go further than a blank page. Start with an example if you like.
02Find a starting point
Use an instant template or request a cloud AI draft when available.
03Make it yours
Check the details, add your own voice, and copy, download, or print your result.
From the field notes.
Make tutorial captions easier to follow
A viewer may be reading while their hands are busy with the next step.
FIELD NOTE / 02SRT and WebVTT: what this converter preserves
Both formats describe timed text, but their syntax and supported features differ.
FIELD NOTE / 03When a subtitle offset fixes sync — and when it does not
A constant delay and gradually increasing drift are different problems.
Before you begin.
Do I need an account?
No account is needed for the tools. Cloud generation has a shared daily allowance and may temporarily be unavailable.
Does anything heavy run on my device?
No AI models are downloaded. Templates and small subtitle files use lightweight browser code. Cloud AI runs on the server.
Are templates and AI drafts different?
Yes. Templates organize your input using fixed structures. The separate AI button sends your input to a cloud model for a new draft. Every result is labeled.
What happens to my input?
Templates stay in your browser. Cloud inputs are sent to Cloudflare when you request generation. This app does not save them. If you enable Remember, your browser stores a draft locally. See the privacy page for provider processing and the short-lived allowance ledger.