CUE Uttaro SongFormer ONNX

INT8 ONNX artifacts prepared for the non-commercial CUE Uttaro desktop application.

Files

muq-int8.onnx: static 30-second MuQ feature encoder
musicfm-int8.onnx: static 30-second MusicFM feature encoder
songformer-int8.onnx: SongFormer structure analysis head
manifest.json: artifact version, byte sizes, and SHA-256 checksums

Input audio is mono PCM at 24 kHz. These files are inference artifacts and do not include the CUE Uttaro application source code.

Intended use

These artifacts are published for personal, research, evaluation, and other non-commercial use. Commercial use is not permitted because the MuQ model weights are released under CC BY-NC 4.0.

Attribution and licenses

This repository contains derived inference artifacts from third-party models. See NOTICE.md before downloading or redistributing them.

SongFormer: https://github.com/ASLP-lab/SongFormer
MuQ: https://github.com/tencent-ailab/MuQ
MusicFM: https://github.com/minzwon/musicfm

No endorsement by the original authors is implied.

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for akineAItech/CUE-Uttaro-SongFormer-ONNX

Base model

ASLP-lab/SongFormer

Quantized

(1)

this model