An AI-Powered Speech Processing Toolkit and Open Source SOTA Pretrained Models, Supporting Speech Enhancement, Separation, and Target Speaker Extraction, etc.

This repo contains the official PyTorch implementation of "Audio Super Resolution in the Spectral Domain" (ICASSP 2023)

Towards High-Quality and Efficient Speech Bandwidth Extension with Parallel Amplitude and Phase Prediction

Official pytorch implementation of the paper: "Catch-A-Waveform: Learning to Generate Audio from a Single Short Example" (NeurIPS 2021)

[ICASSP 2025] "FLowHigh: Towards efficient and high-quality audio super-resolution with single-step flow matching"

Repo for source code of EBEN: Extreme Bandwidth Extension Network

Official implemtation of UniverSR (ICASSP 2026)

Speech to Phoneme, Bandwidth Extension and Speaker Verification using the Vibravox dataset.

Unofficial Pytorch Lightning Implementation of "Real-time Speech Frequency Bandwidth Extension"

The official PyTorch implementation of VM-ASR, a model designed for high-fidelity audio super-resolution.

FastWave is a lightweight diffusion model for general audio super-resolution (any -> 48 kHz). SOTA quality reconstruction metrics with just 1.3 M parameters and ~50 GFLOPs

This repository contains the source code for the implementation of two deep learning models concerning the audio super resolution task.

Towards High-Quality and Efficient Speech Bandwidth Extension with Parallel Amplitude and Phase Prediction

This is an attempt to list interesting audio bandwidth expansion/super-resolution research works.

Low-complexity Real-time Neural Network for Blind Bandwidth Extension of Wideband Speech demo page

Demo of the paper "Robust real-time bandwidth extension for headsets with a microphone conversion target"

Easy to use audio super-resolution and bandwidth extension from CLI or as a Python package.

This project replicates a key-value store across 5 distinct servers to increase bandwidth and ensure availability.