[FEAT][EASY] Add Voice Activity Detection (VAD) to improve overall WER

by mfuntowicz - opened Apr 25

Inference Endpoints Images org Apr 25

Currently the endpoint assumes a naive, 30s window, chunking strategy. Adding a smarter way to split chunk on silent would certainly signficantly improve WER score.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment