Multi-class SVM for stressed speech recognition

被引：0
作者：
Besbes, Salsabil [1 ]
Lachiri, Lied [2 ]
机构：
[1] Univ Tunis El Manar, Natl Sch Engineers Tunis, Signal Image & Informat Technol Lab, BP 37 Le Belvdre, Tunis 1002, Tunisia
[2] Univ Tunis El Manar, Natl Sch Engineers Tunis, BP 37 Le Belvdre, Tunis 1002, Tunisia
来源：
2016 2ND INTERNATIONAL CONFERENCE ON ADVANCED TECHNOLOGIES FOR SIGNAL AND IMAGE PROCESSING (ATSIP) | 2016年
关键词：
speech recognition; multi-class support vector machines; stressed context; SUSAS database; GFCC;
D O I：
暂无
中图分类号：
TM [电工技术]; TN [电子技术、通信技术];
学科分类号：
0808 ; 0809 ;
摘要：
This paper deals with a new automatic stressed recognition system based on kernel classification. We extracted advanced acoustic features from the stressed signals and employed a multi-class Support Vector Machines with different kernels to recognize speech utterances under stress. Gammatone Frequency Cepstral Coefficients are also established. The system implemented is tested using isolated words from SUSAS database with 4 classes: Neutral, Angry, Lombard and Loud. Experimental results show that the best performance is obtained when we use the auditory feature with different descriptors combination but it depends on the type of the kernel used.
引用
收藏
页码：782 / 787
页数：6
相关论文
共 50 条
  • [21] Stressed speech recognition method based on difference subspace combined with dynamic time warping
    Lv, Chengguo
    Zhang, Rubo
    Li, Peihua
    INDUSTRIAL INSTRUMENTATION AND CONTROL SYSTEMS, PTS 1-4, 2013, 241-244 : 1640 - +
  • [22] IMPROVING SPEECH RECOGNITION ON NOISY SPEECH VIA SPEECH ENHANCEMENT WITH MULTI-DISCRIMINATORS CYCLEGAN
    Li, Chia-Yu
    Ngoc Thang Vu
    2021 IEEE AUTOMATIC SPEECH RECOGNITION AND UNDERSTANDING WORKSHOP (ASRU), 2021, : 830 - 836
  • [23] The Multi-level Approach to Speech Corpora Annotation for Automatic Speech Recognition
    Glavatskih, Igor
    Platonova, Tatyana
    Rogozhina, Valeria
    Shirokova, Anna
    Smolina, Anna
    Kotov, Mikhail
    Ovsyannikova, Anna
    Repalov, Sergey
    Zulkarneev, Mikhail
    SPEECH AND COMPUTER (SPECOM 2015), 2015, 9319 : 438 - 445
  • [24] A hybrid SVM/DDBHMM decision fusion modeling for robust continuous digital speech recognition
    Liu, Jingwei
    Wang, Zuoying
    Mao, Xi
    PATTERN RECOGNITION LETTERS, 2007, 28 (08) : 912 - 920
  • [25] Multi-Channel Transformer Transducer for Speech Recognition
    Chang, Feng-Ju
    Radfar, Martin
    Mouchtaris, Athanasios
    Omologo, Maurizio
    INTERSPEECH 2021, 2021, : 296 - 300
  • [26] Multi-stream parameterization for structural speech recognition
    Asakawa, Satoshi
    Minematsu, Nobuaki
    Hirose, Keikichi
    2008 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING, VOLS 1-12, 2008, : 4097 - +
  • [27] Multi-rate HMM quantization for speech recognition
    Vasilache, Marcel
    2008 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING, VOLS 1-12, 2008, : 4341 - 4344
  • [28] Multi resolution discriminative models for subvocalic speech recognition
    Raugas, Mark
    Sridhar, Vivek Kumar Rangarajan
    Prasad, Rohit
    Natarajan, Prem
    11TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION 2010 (INTERSPEECH 2010), VOLS 3 AND 4, 2010, : 2634 - 2637
  • [29] Multi-agent based Arabic speech recognition
    Taha, Muhammad
    Helmy, Tarek
    Alez, Reda Abo
    PROCEEDING OF THE 2007 IEEE/WIC/ACM INTERNATIONAL CONFERENCE ON WEB INTELLIGENCE AND INTELLIGENT AGENT TECHNOLOGY, WORKSHOPS, 2007, : 433 - +
  • [30] Multi-Dialectical Languages Effect on Speech Recognition
    Elfeky, Mohamed G.
    Moreno, Pedro
    Soto, Victor
    1ST INTERNATIONAL CONFERENCE ON NATURAL LANGUAGE AND SPEECH PROCESSING, 2018, 128 : 1 - 8
12345