Text-independent speaker recognition with short utterances

Abstract

This paper presents a new approach to text-independent speaker recognition. The technique, developed to perform with short unknown utterances, models the spectral traits of a speaker with multiple sub-models rather than using a single statistical distribution as done with previous approaches. The recognition is based on the statistical distribution of the distances between the unknown speaker and each of the speaker models. Only frames that are close to one of the speaker's sub-models are considered in the recognition decision, so that speech events not encountered in the training data do not bias the recognition. The technique has been tested on a conversational data base. Models were generated using 100 s of speech from each of 11 male talkers. Unknown speech was obtained one week after the model data. Recognition accuracies of 96%, 87%, and 79% were obtained for unknown speech durations of 10, 5, and 3 s, respectively. The use of multiple sub-models to characterize spectral traits results in improved discrimination between speakers, particularly when short speech segments are recognized. [Work supported by U. S. Air Force, Rome Air Development Center.]

On the Use of Different Speech Representations for Speaker Modeling

“…So far, two approaches have been developed for use of different representations in speaker recognition. One is simply lumping different representations together to constitute a composite representation [2], [7], [28], [33]. To some extent, such an approach improves the performance of speaker recognition but leads to a higher dimensional feature vector of redundancy, given that any representation can individually represent the original speech data.…”

Speaker identification using three signal voice domains during human-robot interaction

“…a voice sample of half a second is long enough to identify the speaker. Similar work, presented by Lite [6], claims an accuracy rate of 79% for 11 users and 3 seconds for each identification. Our system, with 11 users gets an accuracy rate of 86% using 0.5 seconds of voice.…”

Development of speech corpora for speaker recognition research and evaluation in Indian languages

“…Speaker identification has been an important topic of research mostly in American English since the 1960's. The first study on identification and verification of speakers was reported by Kersta (1962aKersta ( , 1962bKersta ( , 1965Kersta ( , 1966Kersta ( , 1970, Kersta and Colangelo (1970), and Bricker and Pruzansky (1966), Li et al (1966), Li and Wrench (1983) respectively. Dialect variation, both zonal and social in origin, has been an important topic of research in American English since 1930's (Cassidy 1993).…”