Evaluation of Spectral and Prosodic Features of Speech Affected by Orthodontic Appliances Using the GMM Classifier

Abstract

The paper describes our experiment with using the Gaussian mixture models (GMM) for classification of speech uttered by a person wearing orthodontic appliances. For the GMM classification, the input feature vectors comprise the basic and the complementary spectral properties as well as the supra-segmental parameters. Dependence of classification correctness on the number of the parameters in the input feature vector and on the computation complexity is also evaluated. In addition, an influence of the initial setting of the parameters for GMM training process was analyzed. Obtained recognition results are compared visually in the form of graphs as well as numerically in the form of tables and confusion matrices for tested sentences uttered using three configurations of orthodontic appliances.K e y w o r d s: spectral and prosodic features of speech, effect of orthodontic appliances, GMM classifier

Storytelling Voice Conversion: Evaluation Experiment Using Gaussian Mixture Models

“…The performed experiment with GMM recognition of applied transformation of the storytelling voices, the analysis and comparison was aimed at investigation of • influence of the initial parameter during the GMM creation on the resulting identification score: the number of applied mixtures of the Gaussian probability density functions N gmix = {4, Table 5. The length of the input feature vector N f eat = 16 was experimentally chosen in correspondence with the obtained results of previous research [23,24]. These feature sets contain the features determined from the spectral envelopes, as well as the supplementary spectral parameters, and the prosodic parameters as described in Table 3.…”

Recognition of Acoustic Signals of Synchronous Motors with the Use of MoFS and Selected Classifiers

“…Engineers and scientists developed a great variety of feature extraction and classification methods [32][33][34][35][36][37][38][39][40][41][42][43][44][45][46][47]. Classification methods are often based on artificial intelligence.…”

Multimodal audio-visual emotion detection

“…Spectral features obtained directly from spectra; spectral characteristics represent phonetic information. Speech systems based on these characteristics have achieved remarkably high degree of accuracy when capturing speech in a clean environment [19]. The absolute values of the fast Fourier transform (FFT) are calculated from the input samples during cepstral analysis, which determines spectral features of voice (following the Hamming window segmentation and weighting).…”