Pitch-robust acoustic feature using single frequency filtering for children's KWS

The pitch and speaking rate are the two significant factors that cause the acoustic mismatch in children's keyword spotting (KWS) system. This paper proposes a pitch-robust acoustic feature based on single fre-quency filtering (SFF) for the development of children's KWS system. In the proposed approach using SFF, the amplitude envelopes (AEs) of the speech data are computed at D-number of selected frequencies separated in Mel scale. The AEs are then averaged over short-time overlapping analysis frames and log-arithmically compressed to represent the D-dimensional feature set per analysis frame, here termed as Mel spaced single frequency average log envelope (MSSF-ALE). By using the proposed MSSF-ALE feature, improved performance is observed for the deep neural network-hidden Markov model-based KWS sys-tem over the standard Mel-frequency cepstral coefficients (MFCC) and MFCC extracted from the smoothed spectra. The relative improvement of 104.44% in term-weighted value (T W V) for children's KWS is ob-served over the MFCC by using MSSF-ALE. The performance of the KWS system is then evaluated with data-augmented training through explicit speaking rate modification of the training data set. The MSSF-ALE provides a relative improvement of 195.94% in T W V over MFCC with the data-augmented training. The MSSF-ALE also results in improved performance than the explored features in noisy test cases. (c) 2021 Elsevier B.V. All rights reserved.