Automatic captions

Generate subtitles without uploading your media

Create timed captions from audio or video with a local Whisper model. Review every word, adjust individual timing and export SRT or WebVTT for editors, players and publishing platforms.

Quick answer

Whisper Online downloads a whisper.cpp model, processes the selected media locally, lets you correct individual timed words, and exports the result without uploading the media file.

Create a transcriptMedia stays on your device

Step by step

The practical workflow

  1. 1

    Select audio or video

    Choose a recording directly from the device. The browser decodes it locally.

  2. 2

    Generate timed words

    Run Tiny, Base or Small through whisper.cpp and WebAssembly.

  3. 3

    Review the captions

    Edit inline or use the detailed list to change text, add tokens and correct timing.

  4. 4

    Export SRT or VTT

    Set caption length and duration limits, then download the format required by the destination.

Automatic first draft, human-controlled result

Speech recognition saves the initial typing, but publishable subtitles still benefit from review. Names, punctuation, specialist language and noisy speech are common correction points.

The editor keeps automatic timing while letting you replace, merge, add or remove words. Caption grouping happens during export, so formatting can change without losing the reviewed transcript.

Subtitles for creators, educators and organizations

Timed captions make spoken video easier to follow without sound and provide an accessibility layer for many viewers. They are useful for tutorials, product demos, interviews, courses, events and social clips.

  • SRT for video editors, platforms and media players
  • WebVTT for HTML video and browser-based players
  • TXT for scripts, notes and repurposing
  • JSON for applications that need individual word timing

What this generator does not do

It does not burn text into the video, translate captions, identify speakers or create animated title compositions. It produces editable transcript data and subtitle files for the next stage of a workflow.

Why use local transcription?

No media upload or account
Editable word-level timing
SRT, VTT, TXT, and JSON export

FAQ

Common questions

Is the automatic subtitle generator free?

Yes. There is no account or per-minute payment. The device performs the transcription work.

Can I change each subtitle word?

Yes. Edit words inline or use Detailed edit to change text and timing, insert tokens and remove rows.

Does it add subtitles to the video?

No. It downloads SRT or VTT files. Import them into an editor, upload them with a video or attach them to a web player.

Which model should I begin with?

Tiny is the safest lightweight starting point. Base can improve the balance of accuracy and size, while Small needs substantially more memory.

Related guides