Unifying General Audio Understanding and Spatial Awareness

🎧 Simulate realistic room impulse responses using PyTorch with GPU support and a clean API for efficient audio processing.

Official implementation of QK-GCC for robust time delay estimation.

RMSAQA-Baseline

This is a repository of beta neural FCASA and heavy-tailed neural FCASA

Official PyTorch implementation for "Iterative Audio Separation with Mixture Consistency via MIMO Model Extension"

This is the official code implementation of BeamFocusNet.

TSE-Omni: multi-cue target speaker extraction code

Official implementation of PathRIR (IWAENC 2026).

Pure C implementation of FastEnhancer. 3.3x faster than the ONNX Runtime graph, same quality.

Official code for "What Do Neural Networks Learn for TDOA Estimation? A Cross-Architecture Probing Study" (Interspeech 2026)

A pure Python implementation of Google's ViSQOL (Virtual Speech Quality Objective Listener) for objective audio/speech quality assessment.

Trying to Replicate PhaseCoder

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

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

A Lightweight Hybrid Dual Channel Speech Enhancement System under Low-SNR Conditions (Interspeech 2025)

[CVPR 2026] FLAC: Few-Shot Acoustic Synthesis with Flow Matching. FLAC enables RIR generation in novel scenes using only one-shot acoustic observation. The repository also provides AGREE, a joint e…

🌋LavaSR: Fast Speech restoration and enhancement

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

Official repository of the "A two-step approach for speech enhancement in low-SNR scenarios using cyclostationary beamforming and DNNs" paper