[](https://github.com/FORARTfe/HyMPS#- "AUDIO section") [](https://github.com/FORARTfe/HyMPS/blob/main/Audio/AI-based.md#-- "AI-based category") [](https://github.com/FORARTfe/HyMPS/blob/main/Audio/AI-Enhancing.md#--- "Enhancers page")
June 6, 2026 · View on GitHub
📁 Aligners - Declippers - Deharshers - Denoisers - Dereverbers - Upscalers - Restorers - Others
Warning
Aligners ⌂
| Repository | Short description | Language | License | Weights | Last commit |
|---|---|---|---|---|---|
| Differentiable APFs | Code for Differentiable Allpass Filters for Phase Response Estimation and Automatic Signal Alignment paper |  | 3 x pth |  | |
| Exploiting Time-Frequency Conformers for Music Audio Enhancement | Implementation of the "Exploiting Time-Frequency Conformers for Music Audio Enhancement" paper |  | 2 x pt |  | |
| StereoPhaseNet | A deep learning project aimed at correcting phase issues in stereo audio recordings |  | NA |  | |
| Synchformer | Source code for "Synchformer: Efficient Synchronization from Sparse Cues" paper |  | 34 x pt |  | |
| RLAutoAlign | Using reinforcement learning to automatically phase align audio |  | NA |  | |
| Audio Enhancement from Multiple Crowdsourced Recordings | Code for Audio Enhancement from Multiple Crowdsourced Recordings: A Simple and Effective Baseline paper |  | NA |  | |
| DNN Auto Align | Aims to align an input signal with a target / reference signal using differentiable all-pass filters |  | NA |  | |
| syncnet_pipeline | SyncNet pipeline to align audio to videos |  | NA |  |
Declippers ⌂
| Repository | Short description | Language | License | Weights | Last commit |
|---|---|---|---|---|---|
| C-OMP-for-Audio-Declipping | Constrained Orthogonal Matching Pursuit for Audio Declipping |  | NA |  | |
| Audio_declipping | Audio De-clipping Via Deep Learning |  | 2 x zip |  | |
| audio-declipper | Audio Declipping with Social Sparsity |  | NA |  | |
| declipnet | Declipping audio with deep learning technique (wavenet) |  | NA |  | |
| not-MIWAE for Audio Declipping | Implementation of "not-MIWAE: Deep Generative Modelling with Missing not at Random Data" paper for audio declipping |  | NA |  |
Deharshers ⌂
| Repository | Short description | Language | License | Weights | Last commit |
|---|---|---|---|---|---|
| pytorch-audio-deharsher | A Deeply-Supervised U-Net CNN for Removing Harsh Resonances from Audio |  | 1 x pth + 1 x onnx |  |
Denoisers ⌂
| Repository | Short description | Language | License | Weights | Last commit |
|---|---|---|---|---|---|
| Sekte | Reducing the Noise in the Audio Signal by Deep Learning Methods |  | 2 x h5 |  | |
| Audio Denoising with Implicit Neural Representations | The research highlights the potential of INR for robust audio denoising, with HyperSound showing superior performance in maintaining audio clarity and quality |  | NA |  | |
| Audio-Denoiser | Audio Denoiser implemented by CNN and GAN |  | NA |  | |
| Music Cleaning Project | The goal of this project is to clean audio files (concert recordings) from noise (conversations, clinking cups, humming) |  | NA |  | |
| ADNAC | Audio Denoising using Neural Audio Codecs |  | NA |  | |
| AI-Denoiser | AI-enhanced noise filtering tool for real-time audio denoising using machine learning and DSP techniques |  | NA |  | |
| Audio-Enhancement-via-ONMF | Code for "On audio enhancement via online non-negative matrix factorization" paper |  | NA |  | |
| audio_denoiser | An AI model to remove the specific noise from the noisy input audio using the essentials of Deep Learning |  | NA |  | |
| AudioDenoiser | Audio Denoiser project made in Python using an ML approach |  | NA |  | |
| Denoise Network | A PyTorch-based implementation of "A Two-Stage U-Net for High-Fidelity Denoising of Historical Recordings" paper |  | NA |  |
Dereverbers ⌂
| Repository | Short description | Language | License | Weights | Last commit |
|---|---|---|---|---|---|
| Learning Audio-Visual Dereverberation | Code for "Learning Audio-Visual Dereverberation" paper |  |  | NA | |
| dereverberation-ml | It aims at using deep learning to retrieve effects from sets of data. |  | NA |  | |
| Dereverb_MetricGAN-U | - |  | NA |  | |
| audio_dereverberation | - |  | NA |  | |
| Audio Reverb Removal with Pytorch | Code to train a custom time-domain autoencoder to dereverb audio |  | NA |  | |
| Dereverberation | dereverberation |  | NA |  |
Upscalers ⌂
| Repository | Short description | Language | License | Weights | Last commit |
|---|---|---|---|---|---|
| Audio Upscaling | Aims to upscale low quality audio into high quality audio using machine learning |  | NA |  | |
| AudioSuperResolution | Generate higher quality audio from a lower one, using Convolutional neural network. |  | 2 x h5 |  | |
| AEROMamba | Official implementation of "An efficient architecture for audio super-resolution using generative adversarial networks and state space models" paper |  | 4 x th |  | |
| Audio Enhancement with Deep Learning | Denoising and upsampling audio files using DNN architectures such as autoencoders, SEGAN and WaveNet |  | 1 x pt |  | |
| Audio Super Resolution Using Neural Networks | Audio super resolution using neural networks |  | NA |  | |
| AudioSR | Versatile Audio Super-resolution at Scale |  | 1 x bin |  | |
| FLowHigh | Towards efficient and high-quality audio super-resolution with single-step flow matching |  | 2 x pt |  | |
| VM-ASR | The official PyTorch implementation of "A Lightweight Dual-Stream U-Net Model for Efficient Audio Super-Resolution" paper |  | 15 x pth |  | |
| Audio Super-Resolution with GANs | Using adversarial learning, it restores lost high-frequency details and natural timbre, producing near-lossless audio for music remastering, streaming, and archival recovery |  | 2 x pth |  | |
| AudioEnhancer | A neural network capable of upsampling audio |  | NA |  | |
| DualCycleGAN | Official implementation of Nonparallel High-Quality Audio Super Resolution with Domain Adaptation and Resampling CycleGANs paper |  | NA |  | |
| Deep Learning for Audio Super-Resolution | Reconstruct high-quality audio from low-quality signals |  | NA |  | |
| Audio Super Resolution | Enhancment of Audio Quality (Bit-Depth and Sampling-Rate) using Deep Learning |  | NA |  | |
| audio-gan | Audio super resolution gan |  | NA |  | |
| audio-super-resolution-with-GAN | audio super resolution with GAN |  | NA |  | |
| EECS-545-Audio-Upscaling | - |  | NA |  | |
| FlashSR | Code for "FlashSR: One-step Versatile Audio Super-resolution via Diffusion Distillation" paper |  | 3 x pth |  | |
| NGAN | Code for the paper "N for Parameter: Efficient Neural Audio Super resolution with GAN" |  | NA |  | |
| NMGAN | A GAN designed to upsample low-quality audio |  | NA |  |
Restorers ⌂
| Repository | Short description | Language | License | Weights | Last commit |
|---|---|---|---|---|---|
| audio-repair | Nearest Neighbor Imputation to Repair Damaged Audio |  | 10 x pkl |  | |
| Audio Restoration for Generative Models | Implements a modular enhancement pipeline designed to improve the perceptual quality of audio generated by text-to-music models |  | 2 x pth |  | |
| AudioPurify | Neural Network Approach to Audio Restoration |  | NA |  | |
| upscalemp3 | Audio Super Resolution model to decompress mp3 into wav file with recovered frequencies built on a "ResUNet" style architecture |  | NA |  | |
| Audio-Undistortion | Audio Signal Processing for Undistorting Audio Clips |  | 7 x pth |  | |
| Audio Restoration Studio | Project for the course of Deep Learning 2025 |  | 4 x pth |  | |
| audio_upscaling | Restore the audio quality of music released during the MP3 era, which killed frequencies above 8 kHz due to the low sampling rate |  | NA |  | |
| Apollo | Music repair method to convert lossy MP3 compressed music to lossless music |  |  | 1 x bin |  |
| AudioDelossifier | Delossify compressed audio (mp3 and others) |  | 6 x h5 |  | |
| BABE | Zero-Shot Blind Audio Bandwidth Extension |  | 4 x pt |  | |
| BABE-2 | An advanced algorithm designed for the enhancement of historical music recordings (official implementation) |  | 6 x pt |  | |
| HarmonyRestorer v1 | Transform any corrupted audio - speech, music, recordings - into professional quality through Self-Organized Neural Networks and advanced source separation |  | NA |  | |
| Music Restoration Studio | Music restoration desktop app focused on restoring old songs by cleaning hiss/hum/clicks, rebalancing stems and comparing original vs enhanced results |  | 2 x ckpt 2 x pth |  | |
| Restoration of Damaged Sound Recordings using Generative Adversarial Networks | Explores the application of Generative Adversarial Networks (GANs) in restoring and enhancing corrupted audio signals |  | NA |  | |
| Stochastic-Restoration-GAN | Stochastic Restoration of Heavily Compressed Musical Audio using Generative Adversarial Networks in PyTorch |  | NA |  | |
| unalias | Can a neural net rectify audio that was undersampled without proper anti-aliasing? |  | NA |  | |
| PyTorch Implementation of Audio-to-Audio Schrodinger Bridges | A diffusion-based audio restoration model for bandwidth extension and inpainting |  | 3 x ckpt |  | |
| GAN_decoded_audio_enhancement | Unofficial PyTorch Implementation of "Audio codec enhancement with generative adversarial networks" paper |  | NA |  | |
| Grosbeak | Audio quality manipulation experiment using deep learning |  | NA |  |
Others ⌂
| Repository | Short description | Language | License | Weights | Last commit |
|---|---|---|---|---|---|
| Bleeding Removal in Music Signals | Neural networks for removal of bleeding in music signals for the sequential application of Music Source Separation |  | NA |  | |
| Audio-Enhancement-GAN | An audio enhancement system which uses GAN |  | NA |  | |
| AudioEnhancer | A model that improve the quality of an audio |  | NA |  | |
| Audio Cycle GAN | GAN improving poorly synthesized audio samples |  | NA |  |