Extreme Audio Time Stretching Using Neural Synthesis

TL;DR

This paper introduces a neural synthesis-based method for audio time stretching that significantly improves quality at large stretching factors, especially for environmental sounds, by better modeling transient and noise components.

Contribution

It presents a novel combination of sines-transients-noise decomposition with WaveNet synthesis to enhance large-factor audio time stretching quality.

Findings

Outperforms four existing TSM algorithms in subjective tests

Provides better transient and noise modeling for large stretching factors

Stereo compatible and suitable for media slow motion applications

Abstract

A deep neural network solution for time-scale modification (TSM) focused on large stretching factors is proposed, targeting environmental sounds. Traditional TSM artifacts such as transient smearing, loss of presence, and phasiness are heavily accentuated and cause poor audio quality when the TSM factor is four or larger. The weakness of established TSM methods, often based on a phase vocoder structure, lies in the poor description and scaling of the transient and noise components, or nuances, of a sound. Our novel solution combines a sines-transients-noise decomposition with an independent WaveNet synthesizer to provide a better description of the noise component and an improve sound quality for large stretching factors. Results of a subjective listening test against four other TSM algorithms are reported, showing the proposed method to be often superior. The proposed method is stereo compatible and has a wide range of applications related to the slow motion of media content.