This paper provides an efficient approach for text-independent speaker identification using the Inverted Mel-frequency Cepstral Coefficients as feature set and Finite Doubly Truncated Gaussian Mixture as Model (FDTGMM). Over the years, Mel-Frequency Cepstral Coefficients (MFCC), modeled on the human auditory system, has been used as a standard acoustic feature set for speech related application...

UNLABELLED BACKGROUND Special consideration has recently been given to cepstral analysis with mel-frequency cepstral coefficients (MFCCs). The aim of this study was to assess the applicability of MFCCs in acoustic analysis for diagnosing occupational dysphonia in comparison to subjective and objective parameters of voice evaluation. MATERIALS AND METHODS The study comprised 2 groups, one of 5...

Source separation by cepstral smoothing of binary masks In this paper, we propose a separation system of speech signals from two convolutive mixtures. The suggested system is based on the combination of blind source separation technique with a time-frequency masking procedure, followed by a smoothing cepstral. Indeed, after separation of signal sources, the estimated binary masks undergo a ceps...

Paper presents an comparative evaluation of features extraction algorithm for a real-time isolated word recognition system based on FPGA. The Mel-frequency cepstral, linear frequency cepstral, linear predictive and their cepstral coefficients were implemented in hardware/software design. The proposed system was investigated in speaker dependent mode for 100 different Lithuanian words. The robus...

Since their inaugural releases in 2007, Google’s Android and Apple’s iOS have grown to dominate the mobile OS market share. Currently, they jointly possess over 99% of global share with being leading Operating System choice worldwide, controlling close 70% Mobile devices enabled exponential growth a plethora applications that play key roles enabling many use cases are pivotal our daily lives. O...

This paper describes a novel approach to monitor a baby and it’s emotion and needs. Feature extraction methods like Magnitude Sum function, Pitch and Energy have been performed to classify the signal. These extraction techniques are proven to be more accurate than the conventional techniques. Although combinations of all three techniques have to be used to achieve 100% accuracy, the computation...

In this paper, we describe a brief overview of the speaker recognition techniques with their processing steps. Speaker recognition has many problems in feature extraction due to the robustness of the speech with noise. Gamma Tone Filter Bank and Wavelet Packet for the speaker recognition have the best performance over the Hidden Markov Model, Mel Frequency Cepstral Coefficient, Dynamic Time War...