Stars
PianoKontext: Expressive Performance Rendering from Deadpan Context
Moshi is a speech-text foundation model and full-duplex spoken dialogue framework. It uses Mimi, a state-of-the-art streaming neural audio codec.
Elucidated Text-To-Audio (ETTA) is a SOTA text-to-audio model with a holistic understanding of the design space and trained with synthetic captions.
Official repository for Aria-MIDI: a MIDI dataset of 1,186,253 transcribed solo-piano recordings.
SyMuRBench: Benchmark for symbolic music representations
UniAudio 2.0: An audio fundation model for text, speech, sound, and music
The open source code of ALMTokenizer2: Towards Low bit-rate and Semantic-rich Audio Tokenizer with Flow-based Scalar Diffusion Transformer Decoder
Official code and pretrained models for Linear Consistency Autoencoders (Lin-CAE), a method to induce linearity in audio autoencoders via data augmentation.
[ICLR 2026] SmartDJ: declarative audio editing with audio langugae model.
Bridging Piano Transcription and Rendering via Disentangled Score Content and Style (ICLR 2026 accepted)
The most powerful local music generation model that outperforms almost all commercial alternatives, supporting Mac, AMD, Intel, and CUDA devices.
C++ Library for tokenizing MIDI files, designed to be compatible with the MIDITok python library
Step-Audio 2 is an end-to-end multi-modal large language model designed for industry-strength audio understanding and speech conversation.
Suno-like music generation studio for HeartMuLa/heartlib - AI-powered music creation with reference audio style transfer
A powerful 3B-parameter, LLM-based Reinforcement Learning audio edit model excels at editing emotion, speaking style, and paralinguistics, and features robust zero-shot text-to-speech
Qwen3-TTS is an open-source series of TTS models developed by the Qwen team at Alibaba Cloud, supporting stable, expressive, and streaming speech generation, free-form voice design, and vivid voice…
HeartMuLa Official Repo: The Most Powerful Open-Source Music Generation Model of 2026
This is the official implementation for the paper "Pianist Transformer: Towards Expressive Piano Performance Rendering via Scalable Self-Supervised Pre-Training".
Multi-lingual large voice generation model, providing inference, training and deployment full-stack ability.
An Industrial-Level Controllable and Efficient Zero-Shot Text-To-Speech System
Amphion (/æmˈfaɪən/) is a toolkit for Audio, Music, and Speech Generation. Its purpose is to support reproducible research and help junior researchers and engineers get started in the field of audi…