A convenient MIDI tokenizer for Deep Learning networks, with multiple encoding strategies

Steerable discovery of neural audio effects

Speech Separation Using an Asynchronous Fully Recurrent Convolutional Neural Network

This is the official implementation of our neural-network-based fast diffuse room impulse response generator (FAST-RIR) for generating room impulse responses (RIRs) for a given acoustic environment.

Code for paper: "Deep Embeddings and Section Fusion Improve Music Segmentation"

PyTorch Implementation of Multi-Singer

The dataset of Speech Recognition

MT3: Multi-Task Multitrack Music Transcription

Phoneme Level Lyrics Alignment and Text-Informed Singing Voice Separation

Automatic lyrics alignment at phoneme or word level with a pre-trained deep neural network.

PyTorch implementation of "FullSubNet: A Full-Band and Sub-Band Fusion Model for Real-Time Single-Channel Speech Enhancement."

Time-stretch audio clips quickly with PyTorch (CUDA supported)! Additional utilities for searching efficient transformations are included.

Efficient Training of Audio Transformers with Patchout

End-to-end OMR system based on deep learning.

Allosaurus is a pretrained universal phone recognizer for more than 2000 languages

Direct design of biquad filter cascades with deep learning by sampling random polynomials.

Pitch-shift audio clips quickly with PyTorch (CUDA supported)! Additional utilities for searching efficient transformations are included.

Max for Live(M4L) Rhythm generator using Variational Autoencoder(VAE)

Implementation of "Slow-Fast Auditory Streams for Audio Recognition, ICASSP, 2021" in PyTorch

Pytorch implementation of the CREPE pitch tracker

Official PyTorch implementation of Contrastive Learning of Musical Representations

Code used in research that led to the paper "An Empirical Evaluation of End-to-End Polyphonic Optical Music Recognition" (ISMIR 2021)

Perform transfer learning for MIR using Jukebox!

code and demo of the ISMIR 2021 paper CollageNet

The codes for the work "Swin-Unet: Unet-like Pure Transformer for Medical Image Segmentation"

PPG-Based Voice Conversion

Official code for UDH: Universal Deep Hiding for Steganography, Watermarking, and Light Field Messaging (Accepted at NeurIPS2020).