Downloads · 30 days
0
mmvap-candor-avcocktail-ft
lggvu/mmvap-candor-avcocktail-ft
mmvap-candor-avcocktail-ft is a machine learning model from lggvu. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as unknown.
Audio-visual early-fusion Voice Activity Projection (VAP) turn-taking model.
Access
Public
Updated Sep 16, 2026
Repo size
87.6 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.ckpt87.6 MB · 100%
At a glance
- License
- unknown
- Access
- Public
- Created
- Sep 16, 2026
- Updated
- Sep 16, 2026
- SHA
- bc506430
Model card
From the Hugging Face model README
MMVAP — Candor -> AVCocktail (fine-tuned)
Audio-visual early-fusion Voice Activity Projection (VAP) turn-taking model.
Training data: Pre-trained on Candor, then fine-tuned on the AVCocktail corpus.
Checkpoint: Epoch 3, val/loss=0.8458 (best checkpoint).
Files
hparams.yamlcheckpoint.ckpt
Source code
This checkpoint was produced with mm-turn-taking.
- License
- unknown
- Created
- Sep 16, 2026
- Updated
- Sep 16, 2026