Privacy policy

Voice to Text is designed to process audio locally in your browser. In the local workflow, your selected file, transcript, and terminology list remain in your browser session.

Local processing

The site does not need an account, database, or upload endpoint to create a local transcript. The speech model may be downloaded from a model host and cached by your browser.

Browser storage

Cached model files and local interface preferences can be removed through your browser settings. Closing the tab removes in-memory audio references.

Advertising

Advertising may be added to informational pages. Ads will not be placed inside the upload, processing, or editor controls.