Effects of a New Voicing Parameter on Pathological Voice Discrimination by SVM
نویسندگان
چکیده
In this paper, we introduce new voicing parameters to describe the speech signal and we study their effects on the classification of disordered voices. These parameters concern the fundamental frequency and the open quotient. The open quotient is defined as the ratio of the open phase by the pitch period. These parameters are calculated using the multiscale product method (MPM). The classification is operated on two pathological databases MAPACI and MEII using SVM classifiers multi-class one-against-all. We consider two types of classifications: a binary classification into normal and pathological voices for female and male speakers and a threecategory classification into edema, nodule and normal voices for the female speakers only. The effects of these new parameters are studied when added to the MFCC coefficients, delta, delta second, and the energy. KeywordsPathological Voices, SVM, MFCC, Open Quotient.
منابع مشابه
The effectiveness of the glottal to noise excitation ratio for the screening of voice disorders.
This paper evaluates the capabilities of the Glottal to Noise Excitation Ratio for the screening of voice disorders. A lot of effort has been made using this parameter to evaluate voice quality, but there do not exist any studies that evaluate the discrimination capabilities of this acoustic parameter to classify between normal and pathological voices, and neither are there any previous studies...
متن کاملA continuous voicing parameter in the frequency domain
In automatic speech analysis, voicing articulatory is often defined as a binary decision: voiced or unvoiced. Linguists agree that this articulatory should be continuous. In this paper, we present a new approach to compute a continuous voicing indicator of a speech frame. This voicing percentage is then evaluated in both a segmentation process and a speech recognition task. Promising results sh...
متن کاملDiscrimination of voice onset time in infancy.
In an effort to determine whether infants can discriminate speech sounds on the basis of a single acoustic cue, timing onset of periodic voicing, two experiments were conducted employing synthetic speech sounds. Naturally produced syllable pairs were also used for comparison. In the first experiment infants evidenced discrimination of a naturally produced /ba/ versus /pa/ pair and a naturally /...
متن کاملClassification of Normal and Pathological Voice Using SVM and RBFNN
The identification and classification of pathological voice are still a challenging area of research in speech processing. Acoustic features of speech are used mainly to discriminate normal voices from pathological voices. This paper explores and compares various classification models to find the ability of acoustic parameters in differentiating normal voices from pathological voices. An attemp...
متن کاملVoice pathology detection and classification using MPEG-7 audio low-level features
In this paper, a new pathological voice detection and pathology classification method based on MPEG-7 audio lowlevel features is proposed. MPEG-7 features are originally used for multimedia indexing, which includes both video and audio. Indexing is related to event detection, and as pathological voice is a separate event than normal voice, we show that MPEG-7 audio low-level features can do ver...
متن کامل