Neural Speech Phase Prediction Based on Parallel Estimation Architecture and Anti-Wrapping Losses

This paper presents a novel speech phase prediction model which predicts wrapped phase spectra directly from amplitude spectra by neural networks. The proposed model is a cascade of a residual convolutional network and a parallel estimation architecture. The parallel estimation architecture is compo...

Full description

Saved in:
Bibliographic Details
Published in:Proceedings of the ... IEEE International Conference on Acoustics, Speech and Signal Processing (1998) pp. 1 - 5
Main Authors: Ai, Yang, Ling, Zhen-Hua
Format: Conference Proceeding
Language:English
Published: IEEE 04.06.2023
Subjects:
anti-wrapping loss
Delays
Estimation
Iterative algorithms
neural network
parallel estimation architecture
phase wrapping
Prediction algorithms
Predictive models
Signal processing algorithms
speech phase prediction
Training
ISSN:2379-190X
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper presents a novel speech phase prediction model which predicts wrapped phase spectra directly from amplitude spectra by neural networks. The proposed model is a cascade of a residual convolutional network and a parallel estimation architecture. The parallel estimation architecture is composed of two parallel linear convolutional layers and a phase calculation formula, imitating the process of calculating the phase spectra from the real and imaginary parts of complex spectra and strictly restricting the predicted phase values to the principal value interval. To avoid the error expansion issue caused by phase wrapping, we design anti-wrapping training losses defined between the predicted wrapped phase spectra and natural ones by activating the instantaneous phase error, group delay error and instantaneous angular frequency error using an anti-wrapping function. Experimental results show that our proposed neural speech phase prediction model outperforms the iterative Griffin-Lim algorithm and other neural network-based method, in terms of both reconstructed speech quality and generation speed.
ISSN:2379-190X
DOI:10.1109/ICASSP49357.2023.10096553