Title :
Spectral and textural features for automatic classification of fricatives
Author :
Frid, Alex ; Lavner, Yizhar
Author_Institution :
Dept. of Comput. Sci., Tel-Hai Coll., Galilee, Israel
Abstract :
Classification of unvoiced fricatives is an important stage in applications such as spoken term detection and audio-video synchronization, and in technologies for the hearing impaired. Due to their acoustic similarity, extraction of multiple features and construction of high-dimensional feature vectors are required for successful classification of these phonemes. In this study two dimensionality reduction algorithms, namely, t-distributed Stochastic Neighbor Embedding (t-SNE) and Sequential Forward Floating Selection (SFFS) were used to obtain a compact representation of the data. A classification stage (kNN or SVM) was then applied, in which we compared the identification rates between the original feature vector and the low-dimensional representation. A total of 1000 unvoiced fricatives (/s/ /sh/ /f/ and /th/) derived from the TIMIT speech database, containing 25000 short frames of 8 ms each, were used for the evaluation. We show that representing the data by a feature vector with as few as 3 dimensions, yields a classification rate of almost 90% which outperforms most of the results obtained in previous studies.
Keywords :
acoustic signal processing; feature extraction; handicapped aids; hearing aids; signal classification; signal representation; spectral analysis; speech recognition; support vector machines; vectors; SFFS; SVM; acoustic similarity; audio-video synchronization; automatic fricative classification; classification stage; compact data representation; dimensionality reduction algorithms; hearing impaired; high-dimensional feature vector construction; identification rates; kNN; low-dimensional representation; multiple feature extraction; phoneme classification; sequential forward floating selection; spectral feature; spoken term detection; t-SNE; t-distributed stochastic neighbor embedding; textural feature; unvoiced fricative classification; Auditory system; Classification algorithms; Feature extraction; Speech; Support vector machine classification; Training; Floating Search; Fricative classification; Support Vector Machine (SVM); dimensionality reduction; t-SNE;
Conference_Titel :
Pacific Voice Conference (PVC), 2014 XXII Annual
Conference_Location :
Krakow
Print_ISBN :
978-1-4799-3699-1
DOI :
10.1109/PVC.2014.6845422