Title :
Unsupervised speaker segmentation in telephone conversations
Author :
Cohen, Arnon ; Lapidus, Vladimir
Author_Institution :
Dept. of Electr. Eng. & Comput. Eng., Ben-Gurion Univ. of the Negev, Beer-Sheva, Israel
Abstract :
Speaker recognition and verification has been used in a variety of commercial, forensic and military applications. The classical problem is that of supervised recognition, in which there is sufficient a priori information on the speakers to be identified. This paper deals with the problem of unsupervised speech segmentation and speaker classification, where no a priori speaker information is available. The algorithm accepts dual-speaker conversation telephone speech data, detects events of simultaneous speakers, and segment the signal by assigning each speech segment to its speaker. Discrete HMM are used, with 12th order cepstral coefficients. Correct recognition rates of more than 90% are demonstrated
Keywords :
cepstral analysis; hidden Markov models; speaker recognition; speech processing; telephony; 12th order cepstral coefficients; commercial applications; correct recognition rates; discrete HMM; dual-speaker conversation telephone speech data; forensic applications; military applications; simultaneous speakers detection; speaker recognition; speaker verification; speech segment; supervised recognition; telephone conversations; unsupervised speaker classification; unsupervised speaker segmentation; unsupervised speech segmentation; Application software; Clustering algorithms; Forensics; Hidden Markov models; Iterative algorithms; Military computing; Signal processing; Speaker recognition; Speech processing; Telephony;
Conference_Titel :
Electrical and Electronics Engineers in Israel, 1996., Nineteenth Convention of
Conference_Location :
Jerusalem
Print_ISBN :
0-7803-3330-6
DOI :
10.1109/EEIS.1996.566903
