MuQ is a large music foundation model pretrained via self-supervised learning using Mel Residual Vector Quantization. It achieves state-of-the-art results on various music information retrieval tasks. The repository also includes MuQ-MuLan, a joint music-text embedding model supporting English and Chinese, with official Python library for feature extraction.
In the Voice, TTS & speech space, MuQ Large Msd Iter takes a focused approach. It focuses on learning rich, general-purpose representations from music audio without labeled data for downstream MIR tasks. It is built as an open-source project for developers. MuQ Large Msd Iter is open source under the MIT license. MuQ Large Msd Iter is available on the web and API.
Tencent AI Lab builds and maintains MuQ Large Msd Iter, and it first shipped in 2024. Development happens publicly on GitHub with 357 stars. Among its 3 catalogued features are Music Representation, Self-Supervised Learning, and Audio Classification.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Same category — not a similarity match