ENPTES

Transcribe video to text with AI, privately

Drop a video you own; GustKit pulls the audio and runs Whisper on your device to produce a timestamped transcript and subtitles.

Runs in your browser Free No sign-up
  1. Choose file
  2. Set options
  3. Get result

Your file never leaves your device. Processed locally with WebAssembly. Nothing is uploaded.

From video to subtitles in three steps

Choose the video, select the spoken language (or leave auto-detect on) and press “Transcribe to text”. When it finishes, download SRT or VTT subtitles for YouTube, Premiere or CapCut, plain TXT, or an editable DOCX.

How accurate is it?

GustKit uses OpenAI's open Whisper models. “Fast” (tiny) is great for clear speech and quick drafts; “Better” (base) is noticeably more accurate on accents and background noise. Clear audio with one speaker gives the best results.

Runs on your GPU when it can

In Chrome and Edge with WebGPU, Whisper runs on your graphics card and is several times faster. Other browsers fall back to WebAssembly on the CPU, which is slower, but the result is the same and still private.

Frequently asked questions

Which languages are supported?

Whisper understands about 99 languages, including English, Portuguese and Spanish. Pick the language for best accuracy, or let it detect automatically.

How long can the video be?

Up to the duration shown in the tool. Transcription runs on your own device, so very long recordings are slow, so split them into parts.

Is my video sent to an AI server?

No. Only the model files are downloaded to your browser; the audio is processed locally and never leaves your device.