Audio-to-Blendshapes Inference

Upload audio or record speech to generate facial blendshapes and head pose data

Audio Input
Record audio directly or upload an audio file (WebM, WAV, MP3, etc.)
OR

Drag and drop an audio file here, or click to select

Ready to process audio
Inference Results
Blendshape and head pose data will appear here after processing

No results yet. Process an audio file to see blendshape data.