Voice Activity Detection (VAD) algorithms based on machine learning techniques have shown competitive results in the area of automatic speech recognition. This paper describes a new approach of VAD based on Support Vector Machines (SVM) for Distributed Speech Recognition (DSR) system. In the proposed scheme, the speech and the non-speech frames are detected from the compressed Mel Frequency Cep...

In this paper, we proposed a weighted discrete K-nearest neighbor (weighted D-KNN) classification algorithm for detecting and evaluating emotion from Mandarin speech. In the experiments of the emotion recognition, Mandarin emotional speech database used contains five basic emotions, including anger, happiness, sadness, boredom and neutral, and the extracted acoustic features are Mel-Frequency C...

Speech Synthesis (SS) and Voice Conversion (VC) presents a genuine risk of attacks for Automatic Speaker Verification (ASV) technology. In this paper, we use our recently proposed unsupervised filterbank learning technique using Convolutional Restricted Boltzmann Machine (ConvRBM) as a frontend feature representation. ConvRBM is trained on training subset of ASV spoof 2015 challenge database. A...

This paper proposes fusion and addition techniques of vocal tract features such as Mel Frequency Cepstral Coefficients (MFCC) and Dynamic Mel Frequency Cepstral Coefficients (DMFCC) in speaker identification. Feature extraction plays an important role as a front end processing block in Speaker Identification (SI) process. Mel frequency features are used to extract the spectral characteristics o...

have been very successful in speech recognition, they have the following two problems: 1) They do not have any physical interpretation, and 2) Liftering of cepstral coefficients, found to be highly useful in the earlier dynamic warping-based speech recognition systems, has no effect in the recognition process when used with continuous observation Gaussian density 4 hidden Markov models. In this...

This work proposes using Wavelet-Packet Cepstral coefficients (WPPCs) as an alternative way to do filter-bank energy-based feature extraction (FE) for automatic speech recognition (ASR). The rich coverage of time-frequency properties of Wavelet Packets (WPs) is used to obtain new sets of acoustic features, in which competitive and better performances are obtained with respect to the widely adop...

This paper describes a medium size Bangla speech corpus preparation and the comparison of the performances of different acoustic features for Bangla word recognition. A small number of speakers are use for most of the Bangla automatic speech recognition (ASR) system, but 40 speakers selected from a wide area of Bangladesh, where Bangla is used as a native language, are involved here. In the exp...

The proposed Missing Feature Linear-Frequency Cepstral Coefficients (MF-LFCC) is a noise robust cepstral feature that transforms both clean and noisy signals into a similar representation. Unlike conventional Missing Feature Techniques, the MF-LFCC does not require the substitution of spectrogram elements (imputation) or classifier modification (marginalization). To improve the noise mask used ...

The most popular speech feature extractor used in automatic speech recognition (ASR) systems today is the mel frequency cepstral coefficient (mfcc) algorithm. Introduced in 1980, the filter bank-based algorithm eventually replaced linear prediction cepstral coefficients (lpcc) as the premier front end, primarily because of mfcc’s superior robustness to additive noise. However, mfcc does not app...

In this paper we propose a new fusion technique, termed Joint Cohort Normalization Fusion, where the information fusion is done prior to the likelihood ratio test in a speaker verification system. The performance of the technique is compared against two popular types of fusion: feature vector concatenation and expert opinion fusion, for fusion of Mel Frequency Cepstral Coefficients (MFCC), MFCC...