Do WaveNets Dream of Acoustic Waves?

by   Kanru Hua, et al.

Various sources have reported the WaveNet deep learning architecture being able to generate high-quality speech, but to our knowledge there haven't been studies on the interpretation or visualization of trained WaveNets. This study investigates the possibility that WaveNet understands speech by unsupervisedly learning an acoustically meaningful latent representation of the speech signals in its receptive field; we also attempt to interpret the mechanism by which the feature extraction is performed. Suggested by singular value decomposition and linear regression analysis on the activations and known acoustic features (e.g. F0), the key findings are (1) activations in the higher layers are highly correlated with spectral features; (2) WaveNet explicitly performs pitch extraction despite being trained to directly predict the next audio sample and (3) for the said feature analysis to take place, the latent signal representation is converted back and forth between baseband and wideband components.


Visualization and Interpretation of Latent Spaces for Controlling Expressive Speech Synthesis through Audio Analysis

The field of Text-to-Speech has experienced huge improvements last years...

Interpreting intermediate convolutional layers of CNNs trained on raw speech

This paper presents a technique to interpret and visualize intermediate ...

WaDeNet: Wavelet Decomposition based CNN for Speech Processing

Existing speech processing systems consist of different modules, individ...

Dereverberation of Autoregressive Envelopes for Far-field Speech Recognition

The task of speech recognition in far-field environments is adversely af...

Speech Denoising with Deep Feature Losses

We present an end-to-end deep learning approach to denoising speech sign...

Inferring Pitch from Coarse Spectral Features

Fundamental frequency (F0) has long been treated as the physical definit...

Please sign up or login with your details

Forgot password? Click here to reset