Phase aware deep neural network for noise robust voice activity detection
Phase aware deep neural network for noise robust voice activity detection
复制标题
DOI：
10.1109/icme.2017.8019414
复制
发表时间：
2017-07
期刊：
影响因子：
--
通讯作者：
Longbiao Wang;Khomdet Phapatanaburi;Zeyan Oo;S. Nakagawa;M. Iwahashi;J. Dang
中科院分区：
文献类型：
--
作者：
Longbiao Wang;Khomdet Phapatanaburi;Zeyan Oo;S. Nakagawa;M. Iwahashi;J. Dang
几乎所有语音活动检测 (VAD) 都会忽略相位信息。为了充分利用原始信号中的信息，本文提出了一种使用幅度和相位信息的深度神经网络（DNN）（即相位感知DNN）来实现更好的VAD性能。梅尔频率倒谱系数（MFCC）、功率归一化倒谱系数（PNCC）、瞬时频率导数（IF）、基带相位差（BPD）和修正群延迟倒谱系数（MGDCC）用作幅度和相位信息。在噪声条件下使用 CENSREC-1-C 数据库对所提出的方法进行了评估。结果表明，相位感知 DNN 的性能显着优于仅使用幅度信息的 DNN。对于基于 DNN 的分类器，等错误率 (EER) 从 MFCC 的 23.70% 降低到联合双幅度和单相位特征（增强 PNCC、MGDCC 和 IF）的 20.43%，再降低到联合双相位和单幅度特征（增强 PNCC、MGDCC 和 BPD）的 19.92%。通过将双相和单相联合特征与双相和单相联合特征相结合，EER 降低至 19.44%。
Phase information is ignored for almost all voice activity detection (VAD). To exploit full information in the original signal, this paper proposes a deep neural network (DNN) using magnitude and phase information (that is, phase aware DNN) to achieve better VAD performance. Mel-frequency cepstral coefficient (MFCC), power-normalized cepstral coefficients (PNCC), instantaneous frequency derivative (IF), baseband phase difference (BPD) and modified group delay cepstral coefficient (MGDCC) are used as magnitude and phase information. The proposed methods were evaluated using CENSREC-1-C database under noise condition. The results show that the phase aware DNN significantly outperforms the DNN using only magnitude information. For DNN-based classifier, the equal error rate (EER) was reduced from 23.70% of MFCC, to 20.43% of joint dual magnitude and single phase features (augmenting PNCC, MGDCC and IF), to 19.92% of joint dual phase and single magnitude feature features (augmenting PNCC, MGDCC and BPD). By combining joint dual magnitude and single phase features with joint dual phase and single magnitude features, the EER was reduced to 19.44%.