WhisperX Studio logo

WhisperX Studio

Basic configuration

Keep it light or open the advanced console when needed.

Step 1Add API key
Stored locally in your browser
Step 2Provide audio file
Audio file URLPublicly accessible audio file link.
Step 3Optional extras
LanguageUse None to enable language detection.
Align outputEnable if you need precise word timings.
Auto detect|Diarization disabled

Results console

Live status, transcripts, and structured JSON output.

idle
Model page
Outputs stream here once complete.

How to use WhisperX Studio

Bring your Replicate token, share an audio source, and let WhisperX handle the transcription workflow.

Quick start
  1. Step 1Add your API token

    Create a token at replicate.com/account/api-tokens and paste it in. It stays in your browser.

  2. Step 2Provide your audio

    Use a public URL or upload to Vercel Blob. Uploads are deleted after completion, and Replicate must be able to reach the URL.

  3. Step 3Tune if needed

    Open Advanced to adjust diarization, alignment, VAD, and language detection.

  4. Step 4Run and review

    Hit Run to start. Track status updates and transcripts in the results console.

Tips and links
Model details

Review pricing, inputs, and docs at replicate.com/victor-upmeet/whisperx.

Good to know
  • Set language to None to enable auto-detection.
  • If you enable diarization, add a HuggingFace token.
  • File uploads are limited to 50 MB.

Advanced configuration

Fine-tune WhisperX input parameters.

Diarization
Required for diarization and stored locally.
Debug