DNN-Based Amplitude and Phase Feature Enhancement for Noise Robust Speaker Identification

Abstract

The importance of the phase information of speech signal is gathering attention. Many researches indicate system combination of the amplitude and phase features is effective for improving speaker recognition performance under noisy environments. On the other hand, speech enhancement approach is taken usually to reduce the influence of noises. However, this approach only enhances the amplitude spectrum, therefor noisy phase spectrum is used for reconstructing the estimated signal. Recent years, DNN based feature enhancement is studied intensively for robust speech processing. This approach is expected to be effective also for phase-based feature. In this paper, we propose feature space enhancement of amplitude and phase features using deep neural network (DNN) for speaker identification. We used mel-frequency cepstral coefficients as an amplitude feature, and modified group delay cepstral coefficients as a phase feature. Simultaneous enhancement of amplitude and phase based feature was effective, and it achieved about 24% relative error reduction comparing with individual feature enhancement.

Within-Sample Variability-Invariant Loss for Robust Speaker Recognition Under Noisy Environments

“…To compensate for these adverse impacts, various approaches have been proposed at different stages of the ASV systems. At the signal level, DNN based speech or feature enhancement [4,5,6,7] has been investigated for ASV under complex environment. At the feature level, feature normalization techniques [8] and noise-robust features such as power-normalized cepstral coefficients (PNCC) [9] have also been applied to ASV systems.…”

Far-Field End-to-End Text-Dependent Speaker Verification Based on Mixed Training Data with Transfer Learning and Enrollment Data Augmentation

“…At the signal level, linear prediction inverse modulation transfer function [7] and weighted prediction error (WPE) [8,9] methods have been used for dereverberation. DNN based denoising methods for single-channel speech enhancement [10,11,12,13] and beamforming for multi-channel speech enhancement [8,14,15] have also been explored for ASV system under complex environments. At the feature level, sub-band Hilbert envelopes based features [16,17,18], warped minimum variance distortionless response (MVDR) cepstral coefficients [19], blind spectral weighting (BSW) based features [17], power-normalized cepstral coefficients (PNCC) [20] and DNN bottleneck features [21] have been applied to ASV system to suppress the adverse impacts of reverberation and noise.…”

The DKU System for the Speaker Recognition Task of the 2019 VOiCES from a Distance Challenge

“…We also thank Weixiang Hu, Yu Lu, Zexin Liu and Lei Miao from Huawei Digital Technologies Co., Ltd, China. been proposed at different stages of the speaker recognition system. At the signal level, dereverberation [6], denoising [7,8,9,10], and beamforming [11,12] can be used for speech enhancement. At feature level, sub-band Hilbert envelopes based features [13,14], warped minimum variance distortionless response (MVDR) cepstral coefficients [15], blind spectral weighting (BSW) based features [16] have been applied to ASV system to suppress the adverse impacts of reverberation and noise.…”