[ICLR 2026] SmartDJ: declarative audio editing with audio langugae model.

Python Toolbox for Feedack Delay Networks (FDNs)

Differentiable framework-agnostic audio DSP with a functional graph API, a modern CLI, and a hard-real-time engine

A controllable, end-to-end API for soundscape synthesis across ray-traced & real-world measured acoustics

Official implementation of “Learning Where to Split: Dynamic Band Splitting for Audio Source Separation”

Unifying General Audio Understanding and Spatial Awareness

Reference-aware automatic speech evaluation toolkit

Official implementation of PhysWave [EMNLP'26 Main]

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

SALMONN family: A suite of advanced multi-modal LLMs

This paper has been accepted for publication in The Fourteenth International Conference on Learning Representations (ICLR) 2026.

Spherical Harmonics ROOM, an open-source Python library for room acoustics simulation using Ambisonics, https://arxiv.org/abs/2603.27342, installable via {pip install pyshroom}.

An STFT/iSTFT for PyTorch.

A training code template for DNN-based speech enhancement.

A repository and benchmark for online test-time adaptation.

🌋LavaSR: Fast Speech restoration and enhancement

This repository contains the training, inference, evaluation code for SpeechLLM models and details about the model releases on huggingface.

Learning audio concepts from natural language supervision

The Open Source Code of UniAudio

Code for ICML 2024 paper (Oral) — Test-Time Model Adaptation with Only Forward Passes

MRSAudio: A Large-Scale Multimodal Recorded Spatial Audio Dataset with Refined Annotations

The hub for audio AI research: papers, open models, benchmarks & datasets across audio LLMs, speech recognition, TTS, music & audio generation.

AudioBench: A Universal Benchmark for Audio Large Language Models

Real-time Speech-Text Foundation Model Toolkit (wip)