Skip to content
Latest Effects
Captions / Transcribe speech
  • local
  • Local
  • Tools

Local Whisper

Transcribe multilingual audio locally with the open Whisper model.

Explore Local Whisper inputs and outputs. Transcribing a clip on your own device and preparing timed words for editable captions.

  • Transcribe speech

Local Whisper pricing example

$0.00 Estimated per run

Estimate checked . This example uses the configuration shown above. Inputs and settings can change the price; check the estimate for your own run before generating. Project storage is billed separately.

How AI usage and project pricing work →

Curated output references.

Current API configuration

What goes in. What comes back.

The model defines every slot it reads and every result it returns. Required inputs must be present before a run can begin.

Inputs

Prompts and source material

Source

Audio and Video · Required

Outputs

Files and structured results

Captions

Metadata

Available controls

Configure the run.

Control Available values Requirement
Language en, zh, de, es, ru, ko, fr, ja, pt, tr, pl, ca, nl, ar, sv, it, id, hi, fi, vi, he, uk, el, ms, cs, ro, da, hu, ta, no, th, ur, hr, bg, lt, la, mi, ml, cy, sk, te, fa, lv, bn, sr, az, sl, kn, et, mk, br, eu, is, hy, ne, mn, bs, kk, sq, sw, gl, mr, pa, si, km, sn, yo, so, af, oc, ka, be, tg, sd, gu, am, yi, lo, uz, fo, ht, ps, tk, nn, mt, sa, lb, my, bo, tl, mg, as, tt, haw, ln, ha, ba, jw, and su Optional

Works especially well for

  • Transcribing a clip on your own device and preparing timed words for editable captions.

Know the trade-offs

  • Requires compatible WebCodecs and an initial model download. Noise, names, and overlapping speech can cause errors; generated caption styling is separate from transcription accuracy.

Configure the model.
Keep the result in the cut.

The generation returns to the scene, asset folder and timeline it belongs to.