On the amount of speech data necessary for successful speaker identification

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23520%2F03%3A00000154" target="_blank" >RIV/49777513:23520/03:00000154 - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—

Result language
angličtina
Original language name
On the amount of speech data necessary for successful speaker identification
Original language description
The paper deals with the dependence between the speaker identification performance and the amount of test data. Three speaker identification procedures based on hidden Markov models (HMMs) of phonemes are presented here. One, which is quite commonly us ed in the speaker recognition systems based on HMMs, uses the likelihood of the whole utterance for speaker identification. The other two that are proposed in this paper are based on the majority voting rule. The experiments were performed for two diff erent situations: either both training and test data were obtained from the same channel, or they were obtained from different channels. All experiments show that the proposed speaker identification procedure based on the majority voting rule for sequen ces of phonemes allows us to reduce the amount of test data necessary for successful speaker identification.
Czech name
—
Czech description
—

Project
<a href="/en/project/GA102%2F02%2F0124" target="_blank" >GA102/02/0124: Voice technologies for support of information society</a><br>
Continuities
Z - Vyzkumny zamer (s odkazem do CEZ)

Publication year
2003
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Similar results(10)