Yayın: Principal component based classification for text-independent speaker identification
Tarih
Kurum Yazarları
Hanilçi, Cemal
Ertaş, Figen
Yazarlar
Danışman
Dil
Türü
Yayıncı:
IEEE
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Özet
Classification based on Principal Component analysis has recently appeared in the literature in application to text-independent speaker identification. However, results have been reported for only clean speech data. In this paper, we evaluate the performance of principal component classifier for text-independent speaker identification on telephone speech. We then improve its identification performance using a Vector Quantization classifier in combination, through fusion of classifier scores. An identification rate of 78.27% has been obtained on the NTIMIT database, which is well above the best identification rate ever reported in the literature obtained by using only one type of feature set.
Açıklama
Bu çalışma, 02-04 Eylül 2010 tarihleri arasında Famagusta[Kuzey Kıbrıs Türk Cumhuriyeti]’da düzenlenen 5. International Conference on Soft Computing, Computing with Words and Perceptions in System Analysis, Decision and Control’da bildiri olarak sunulmuştur.
Kaynak:
Anahtar Kelimeler:
Konusu
Computer science, Engineering, Classifiers, Identification (control systems), Independent component analysis, Loudspeakers, Soft computing, Speech recognition, Systems analysis, Text processing, Vector quantization, Clean speech, Feature sets, Fusion of classifiers, Identification rates, Principal component classifiers, Principal components, Telephone speech, Text-independent speaker identification, Principal component analysis
Alıntı
Hanilçi, C. ve Ertaş, F. (2010). "Principal component based classification for text-independent speaker identification". ICSCCW 2009 - 5th International Conference on Soft Computing, Computing with Words and Perceptions in System Analysis, Decision and Control, 39-42.