Different types of cepstral analysis are compared in the context of a percussion instrument classification external for Pd. For raw cepstrum, mel frequency cepstrum, DCT-based cepstrum, and bark frequency cepstrum, various parameter settings are applied to a standardized test. Significant score improvement can be seen when moving from cepstrum to mel cepstrum, and further improvement is achieve...

Automatic voice pathology detection enables objective assessment of pathologies that affect the production mechanism. Detection systems have been developed using traditional pipeline approach (consisting feature extraction part and part) modern deep learning -based end-to-end approach. Due to lack vast amounts training data in study area pathological voice, former is still a valid choice. In ex...

In this paper we propose a multichannel feature compensation approach for automatic speech recognition in reverberant and noisy environments. The proposed technique propagates the posterior of the clean signal estimated by a multichannel Wiener filter in short-time Fourier transform (STFT) domain into Mel-frequency cepstrum coefficients (MFCC) domain. The multichannel Wiener filter reduces both...

Abstract In this paper, general regression neural network (GRNN) with the input feature of Mel-frequency cepstrum coefficient (MFCC) is employed to automatically recognize calls leopard, ross, and weddell seals widely overlapping living areas. As a feedforward network, GRNN has only one parameter, i.e., spread factor. The recognition performance can be greatly improved by determining factor bas...

In this study, an architecture called Convolutional Long Short-term memory deep neural network (CLDNN) based on learning, which has not been used before in field, is for music genre classification. addition, a new Turkish Music Database consisting of 200 belonging to various genres created. The classification performance the proposed and commonly machine learning methods evaluated database. fea...

We propose a new front-end that reflects some aspects of auditory nerve response. Namely, the pattern of synchrony responses observed over auditory nerve fibers associated with F0, F1 and F2 of voiced sounds. The main goal is to get a set of features, which represents those frequency trajectories. These features should be less sensitive to adverse environmental conditions than mel-cepstrum or F...

Akfract-Several parametric representations of the acoustic signal were compared with regard to word recognition performance in a syllable-oriented continuous speech recognition system. The vocabulary included many phonetically similar monosyllabic words, therefore the emphasis was on the ability to retain phonetically significant acoustic information in the face of syntactic and dura...

Akfract-Several parametric representations of the acoustic signal were compared with regard to word recognition performance in a syllable-oriented continuous speech recognition system. The vocabulary included many phonetically similar monosyllabic words, therefore the emphasis was on the ability to retain phonetically significant acoustic information in the face of syntactic and dura...

In this paper a text-dependent speaker recognition method is presented by combining Mel frequency cepstrum coefficients (MFCC) and Euclidean distance. The robustness of this speaker identification method for different speaking language is analyzed in this paper. The speaker identification algorithm using English and Hindi Indian voice database (IVD) which contains sentences of data spoken is ac...