Automatic speaker, age-group and gender identification from children's speech. (July 2018)

• Record Type: Journal Article
• Authors: Safavi, Saeid; Russell, Martin; Jančovič, Peter
• Is Part Of: Computer speech & language. Volume 50(2018)
• Page Start: 141
• Page End: 156
• Publication Date: 2018-07
• Subjects: Speaker recognition -- Gender identification -- Age-group identification -- Children's speech -- Gaussian Mixture Model (GMM) -- Support Vector Machine (SVM) -- i-vector; Speech processing systems -- Periodicals; Automatic speech recognition -- Periodicals; Computers -- Periodicals; Linguistics -- Periodicals; Speech-Language Pathology -- Periodicals; Traitement automatique de la parole -- Périodiques; Reconnaissance automatique de la parole -- Périodiques; Automatic speech recognition; Speech processing systems; Electronic journals; Periodicals; 006.454
• DOI: 10.1016/j.csl.2018.01.001
• Languages: English
• ISSNs: 0885-2308

Abstract

Highlights: Systematic study of acoustic approaches to speaker, gender and age-group identification from speech for children. Analysis of the utility of frequency sub-bands for speaker, gender and age-group identification from children's speech. Application of GMM-UBM, GMM-SVM and i-vectors to speaker, gender and age-group identification from children's speech. Analysis of the effects of the onset of puberty in boys on automatic speaker, gender and age-group identification. Abstract: A speech signal contains important paralinguistic information, such as the identity, age, gender, language, accent, and the emotional state of the speaker. Automatic recognition of these types of information in adults' speech has received considerable attention, however there has been little work on children's speech. This paper focuses on speaker, gender, and age-group recognition from children's speech. The performances of several classification methods are compared, including Gaussian Mixture Model–Universal Background Model (GMM–UBM), GMM–Support Vector Machine (GMM–SVM) and i-vector based approaches. For speaker recognition, error rate decreases as age increases, as one might expect. However for gender and age-group recognition the effect of age is more complex due mainly to consequences of the onset of puberty. Finally, the utility of different frequency bands for speaker, age-group and gender recognition from children's speech is assessed.