End-to-end music source separation: is it possible in the waveform domain? (2018-10-29T00:00:00.000000Z)

TL;DR

A Wavenet-based model is proposed and Wave-U-Net can outperform DeepConvSep, a recent spectrogram-based deep learning model, and the results confirm that waveform-based models can perform similarly (if not better) than a spectrogram/deep learning model.

Abstract

Most of the currently successful source separation techniques use the magnitude spectrogram as input, and are therefore by default omitting part of the signal: the phase. To avoid omitting potentially useful information, we study the viability of using end-to-end models for music source separation --- which take into account all the information available in the raw audio signal, including the phase. Although during the last decades end-to-end music source separation has been considered almost unattainable, our results confirm that waveform-based models can perform similarly (if not better) than a spectrogram-based deep learning model. Namely: a Wavenet-based model we propose and Wave-U-Net can outperform DeepConvSep, a recent spectrogram-based deep learning model.

Authors

Xavier Serra

14 papers

Jordi Pons

6 papers

Francesc Lluís

1 papers

TL;DR

Abstract

Authors

References34 items

End-to-end Networks for Supervised Single-channel Speech Separation

Phasebook and Friends: Leveraging Discrete Representations for Source Separation

Wave-U-Net: A Multi-Scale Neural Network for End-to-End Audio Source Separation

Randomly Weighted CNNs for (Music) Audio Classification

The 2018 Signal Separation Evaluation Campaign

Raw Multi-Channel Audio Source Separation using Multi- Resolution Convolutional Auto-Encoders

Natural TTS Synthesis by Conditioning Wavenet on MEL Spectrogram Predictions

Parallel WaveNet: Fast High-Fidelity Speech Synthesis

End-to-end Learning for Music Audio Tagging at Scale

Does Phase Matter For Monaural Source Separation?

TaSNet: Time-Domain Audio Separation Network for Real-Time, Single-Channel Speech Separation

Singing Voice Separation with Deep U-Net Convolutional Networks

A Wavenet for Speech Denoising

End-To-End Source Separation With Adaptive Front-Ends

Timbre analysis of music audio signals with convolutional neural networks

Designing efficient architectures for modeling temporal features with convolutional neural networks

Monoaural Audio Source Separation Using Deep Convolutional Neural Networks

The 2016 Signal Separation Evaluation Campaign

WaveNet: A Generative Model for Raw Audio

U-Net: Convolutional Networks for Biomedical Image Segmentation

On automatic drum transcription using non-negative matrix deconvolution and itakura saito divergence

Singing-Voice Separation from Monaural Recordings using Deep Recurrent Neural Networks

A fast Griffin-Lim algorithm

Complex NMF: A new sparse representation for acoustic signals

Unsupervised learning of sparse and shift-invariant decompositions of polyphonic music

Extracting Sound Objects by Independent Subspace Analysis

Separation of harmonic sound sources using sinusoidal modeling

Independent component analysis: algorithms and applications

Sparse coding with an overcomplete basis set: A strategy employed by V1?

Nal Kalchbrenner, Andrew Senior, and Koray Kavukcuoglu

Unsupervised Learning Methods for Source Separation in Monaural Music Signals

BSS_EVAL Toolbox User Guide -- Revision 2.0

A Maximum Likelihood Approach to Single-channel Source Separation

Algorithms for Non-negative Matrix Factorization

Field of Study

Journal Information

Name

Volume

Venue Information

Name

Type

URL

Alternate Names