Pipeline for Realtime Music Enhancement

Code for "Separate-and-detect: Unified drum transcription and stem generation via latent diffusion"

Unified registry of 85 audio effects across 7 libraries (audiomentations, sox, torchaudio, scipy, librosa, pyloudnorm, numpy), with a clean chain-application API inspired by pedalboard.

Official implementation of the ISMIR 2026 paper StemFX: learning mixing style representations via autoregressive FX chain prediction on source-separated stems.

MuScriptor is a multi-instrument music transcription model developed by Kyutai and Mirelo.

(ICCV'25) TF-TI2I: Training-Free Text-and-Image-to-Image Generation via Multi-Modal Implicit-Context Learning in Text-to-Image Models (Augment your T2I models with arbitrary number of images as ref…

Implementation of HS-TasNet, "Real-time Low-latency Music Source Separation using Hybrid Spectrogram-TasNet"

Stemdeck is an modern stem extraction platform for musicians,producers and hobbyists, designed to isolate vocals, drums, bass, piano and guitar for practice, transcription, remixing, and creative a…

DEMON: Diffusion Engine for Musical Orchestrated Noise

[KDD 2026] Official implementation of AnchorSteer: Self-Discovered Concept Injection for Structure-Preserving Music Editing

Realtime audio generation model using Flow Matching DiT on CoDiCodec latents.

Official code for "MIDI-Informed Singing Accompaniment Generation in a Compositional Song Pipeline"

The official repository TimeAudio, a comprehensive framework that incorporates fine-grained acoustic cues into LALMs with enhanced module designs and a specially curated dataset.

[ACMMM 26] Codes Repository for Paper "LaDA-Band: Language Diffusion Models for Vocal-to-Accompaniment Generation"

Fully quantized Neural Networks for Audio Source Separation

The OMAPL Corpus is a comprehensive Mandarin pronunciation assessment dataset comprising 82 Mandarin utterances from native speakers and 1,768 Mandarin utterances from French L1 speakers learning M…