Multitaper MFCC Features for Acoustic Stress Recognition from Speech

Abstract

Ameliorating the performances of speech recognition system is a challenging problem interesting recent researchers. In this paper, we compare two extraction methods of Mel Frequency Cepstral Coefficients used to represent stressed speech utterances in order to obtain best performances. The first method known as traditional is based on single window (taper) generally the Hamming window and the second one is a novel technique developed with multitapers instead of a single taper. The extracted features are then classified using the multiclass Support Vector Machines. Experimental results on the SUSAS database have shown that the multitaper MFCC features outperform the conventional MFCCs.

Authors and Affiliations

Salsabil Besbes, Zied Lachiri

Keywords

Related Articles

The User Behavior Analysis Based on Text Messages Using Parafac and Block Term Decomposition

Tensor decompositions represent a start for big data analysis and a start in reduction of dimensionality, object detection, clustering and so on. This paper presents a method to study the behavior of users in the online...

Design, Modeling and Energy Management of a PEM Fuel Cell / Supercapacitor Hybrid Vehicle

This work concerns the study and the modeling of hybrid Proton Exchange Membrane (PEM) Fuel Cell electric vehicle. In fact, the paper deals with the model description of the powertrain which includes two energy sources:...

Cybercrime in Morocco

Cybercrime encompasses all illegal actions and facts that target cyberspaces and cause enormous economic and financial damages to organizations and individuals. A cyberspace is essentially composed of digital information...

Wireless Sensor Networks for Road Traffic Monitoring

Wireless Sensor Networks (WSNs) consist of large number of sensor nodes. Each node is empowered by a com-munication interface that is mainly characterized by low power, short transmission distance and minimal data rate s...

 Throughput Analysis of Ieee802.11b Wireless Lan With One Access Point Using Opnet Simulator

 This paper analyzes the throughput performance of IEEE 802.11b Wireless Local Area Network (WLAN) with one access point. The IEEE 802.11b is a wireless protocol standard. In this paper, a wireless network was estab...

Download PDF file
  • EP ID EP251116
  • DOI 10.14569/IJACSA.2017.080361
  • Views 101
  • Downloads 0

How To Cite

Salsabil Besbes, Zied Lachiri (2017). Multitaper MFCC Features for Acoustic Stress Recognition from Speech. International Journal of Advanced Computer Science & Applications, 8(3), 446-451. https://europub.co.uk./articles/-A-251116