Official repository for GraFPrint: an audio identification framework based on graph neural networks.

Multi-Scale Neural Audio Codec (SNAC) compresses audio into discrete codes at a low bitrate

Official repo for DiscoDiff: Coarse-to-Fine Text-to-Music Latent Diffusion presented at ICASSP 2025

Contrastive learning of music representations using playlist data

State-of-the-art audio codec with 90x compression factor. Supports 44.1kHz, 24kHz, and 16kHz mono/stereo audio.

[PyTorch] Minimal codebase for MusicGen models

Source code for "Synchformer: Efficient Synchronization from Sparse Cues" (ICASSP 2024)

Dify is an open-source LLM app development platform. Dify's intuitive interface combines AI workflow, RAG pipeline, agent capabilities, model management, observability features and more, letting yo…

Silero VAD: pre-trained enterprise-grade Voice Activity Detector

Supplementary code for the experiments described in the 2021 ISMIR submission: Leveraging Hierarchical Structures for Few Shot Musical Instrument Recognition.

Utility tools for the "Divide and Remaster" dataset, introduced as part of the Cocktail Fork problem paper: https://arxiv.org/abs/2110.09958

Banquet: A Stem-Agnostic Single-Decoder System for Music Source Separation Beyond Four Stems

The official code repo of "HTS-AT: A Hierarchical Token-Semantic Audio Transformer for Sound Classification and Detection"

Server for the MusicBrainz project (website, API, database tools)

Source code for training models and using the hyperbolic interface proposed in our ICASSP 2023 paper, “Hyperbolic Audio Source Separation.”

Code for the ICASSP-2021 paper: Continuous Speech Separation with Conformer.

AAAI 2025: Codec Does Matter: Exploring the Semantic Shortcoming of Codec for Audio Language Model

Towards Open-source GPT-4o with Vision, Speech and Duplex Capabilities。

Svr-fiR Lossless Audio codec

Official implementation of the paper "BigCodec: Pushing the Limits of Low-Bitrate Neural Speech Codec"

Ultra-low-bitrate Speech Codec for Speech Language Modeling Applications

Teaching material for the course "Deep Learning for Music Analysis and Generation" I taught at National Taiwan University (2023 Fall)

PyTorch implementation of WaveFit [2022, Google] which is one of SOTA lightweight/fast speech vocoders.

2021 ISMIR tutorial - music classification

A Music Recommendation System Using the Famous Spotify Million Playlist Dataset

Riemannian Adaptive Optimization Methods with pytorch optim

Schedule-Free Optimization in PyTorch

The Harmonix Set: Beats, Downbeats, and Structural Annotations for Pop Music