Downloads · 30 days
0
dance-baby-dance-mint
neocollab/dance-baby-dance-mint
dance-baby-dance-mint is a audio-to-audio model from neocollab. Use it for the audio-to-audio task on the model card, and read the license before you ship it in a product.
Repository: neocollab/dance-baby-dance-mint Model: mintfactb32v322021-05-06 Checkpoint: ckpt-214501 Framework: TensorFlow 2.5 Python Version: 3.7.16
Access
Public
Updated Aug 9, 2026
Repo size
1.4 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.data-00000-of-000011.4 GB · 100%
At a glance
- Task
- Audio-to-Audio
- Access
- Public
- Created
- Aug 9, 2026
- Updated
- Aug 9, 2026
- SHA
- 95397fb9
Model card
From the Hugging Face model README
Dance Baby Dance - mint-main AI Choreographer
Repository: neocollab/dance-baby-dance-mint
Model: mint_fact_b32_v3_2_2021-05-06
Checkpoint: ckpt-214501
Framework: TensorFlow 2.5
Python Version: 3.7.16
Overview
This repository contains the model checkpoint and configuration assets for the mint-main AI Choreography engine, which is utilized by the Dance Baby Dance platform.
It takes audio features as input and outputs fully synchronized 3D SMPL motion data.
Important Note: This repository strictly contains model checkpoint/configuration assets and is NOT itself the inference API. The Dance Baby Dance platform handles audio processing, feature extraction, autoregressive inference, and local retargeting pipelines elsewhere.
Input
Audio features generated from music/audio tracks (librosa mel-spectrograms/MFCCs).
Output
SMPL motion.
- poses:
[frames, 24, 3](Axis-angle rotations for 24 SMPL joints) - trans:
[frames, 3](Global translation vector) - fps: 60
Usage
These assets must be loaded using the legacy mint-main codebase with Python 3.7 and TensorFlow 2.5.
- Task
- Audio-to-Audio
- Created
- Aug 9, 2026
- Updated
- Aug 9, 2026