Combining short-term cepstral and long-term prosodic features for automatic recognition of speaker age

The most successful systems in previous comparative studies on speaker age recognition used short-term cepstral features modeled with Gaussian Mixture Models (GMMs) or applied multiple phone recognizers trained with the data of speakers of the respective class. Acoustic analyses, however, indicate that certain features such as pitch extracted from a longer span of speech correlate clearly with the speaker age although the systems based on those features have been inferior to the be-fore mentioned approaches. In this paper, three novel systems combining short-term cepstral features and long-term features for speaker age recognition are compared to each other. A sys-tem combining GMMs using frame-based MFCCs and Support-Vector-Machines using long-term pitch performs best. The re-sults indicate that the combination of the two feature types is a promising approach, which corresponds to findings in related fields like speaker recognition. Index Terms: speaker classifi-cation, age recognition, GMM, SVM 1