Comparison of speech parameterization techniques for the classification of speech disfluencies

Fook, Chong Yen; Hariharan, M.; Chee, Lim Sin; Yaacob, Sazali; Adom, Abdul Hamid

doi:10.3906/elk-1112-84

Cited by 28 publications

(12 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In 2009, Ravikumar et al [24] attempted to improve their previous results [23] through the SVM with MFCC features. Yen Fook et al [10] presented a comparison of three feature extraction methods such as the MFCC, linear predictive coding (LPC) and perceptual linear predictive (PLP) analysis for the classification of repetition and prolongation dysfluencies. Three different classifiers (k-NN, LDA, and SVM) were employed and compared in the study.…”

Section: Related Workmentioning

confidence: 99%

“…Over the past several years, considerable studies have been carried out on automatic detection and classification of stuttering dysfluencies [1-3, 10, 23, 24, 32, 34, 35]. Most of them focus on acoustic analysis, parametric and nonparametric feature extraction, and statistical approaches [10,24,32,34,35].…”

Section: Related Workmentioning

confidence: 99%

“…Further, SVM classifiers were used in dysfluency recognition by authors of [10,24]. In 2009, Ravikumar et al [24] attempted to improve their previous results [23] through the SVM with MFCC features.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Gaussian Mixture Model Based Classification of Stuttering Dysfluencies

Mahesha¹,

Vinod²

2016

Journal of Intelligent Systems

View full text Add to dashboard Cite

The classification of dysfluencies is one of the important steps in objective measurement of stuttering disorder. In this work, the focus is on investigating the applicability of automatic speaker recognition (ASR) method for stuttering dysfluency recognition. The system designed for this particular task relies on the Gaussian mixture model (GMM), which is the most widely used probabilistic modeling technique in ASR. The GMM parameters are estimated from Mel frequency cepstral coefficients (MFCCs). This statistical speakermodeling technique represents the fundamental characteristic sounds of speech signal. Using this model, we build a dysfluency recognizer that is capable of recognizing dysfluencies irrespective of a person as well as what is being said. The performance of the system is evaluated for different types of dysfluencies such as syllable repetition, word repetition, prolongation, and interjection using speech samples from the University College London Archive of Stuttered Speech (UCLASS).

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Gaussian Mixture Model Based Classification of Stuttering Dysfluencies

Mahesha¹,

Vinod²

2016

Journal of Intelligent Systems

View full text Add to dashboard Cite

show abstract

“…Formant description depends on order of filter p [22] , it is taken as 30 [28]. Low order cepstral coefficients are sensitive to the overall spectral slope and the high-order cepstral coefficients are sensitive to noise and other forms of noise like variability [25].…”

Section: B Mel-frequency Cepstral Coefficients (Mfcc)mentioning

confidence: 99%

“…The Mel scale can be converted from the frequency scale using (2) given below. Each overlapping mel filter can be viewed as a histogram bin in the frequency domain [25]. Each filter in the bank is multiplied by the spectrum so that only one single value of magnitude per filter is returned.…”

Section: B Mel-frequency Cepstral Coefficients (Mfcc)mentioning

confidence: 99%

Feature enhancement for classifier optimization and dimensionality reduction

Shilaskar¹,

Ghatol²

2014

2014 Annual IEEE India Conference (INDICON)

View full text Add to dashboard Cite

Voice is important for professionals like speakers, teachers, actors, singers and it is the important tool for communication. Laryngeal pathologies induce perturbations in the speech signal. Speech signal is discriminated as pathological or healthy based on roughness -breathiness -hoarseness (RBH) in the quality of signal. In recent years pattern recognition along with various signal processing techniques has emerged as an effective non invasive tool for diagnosis of pathological condition. Signal processing techniques tend to generate large number of features representing the signal. Automatic feature reduction techniques are vital in identifying the relevant features and eliminating the redundant ones. We extract features from speech signal using the acoustic analysis. Features are enhanced by alleviating gender bias. Periodic variations in the signal are captured using statistical techniques. We investigate intelligent system to generate reduced feature subset with improvement in diagnostic performance.

show abstract