Pregled bibliografske jedinice broj: 199568
Speaker Recognition Method Combining FFT, Wavelet Functions And Neural Networks
Speaker Recognition Method Combining FFT, Wavelet Functions And Neural Networks // The Proceedings of the AES 26th International Conference / Pritts, Roy (ur.).
Denver (CO), 2005. str. 47-50 (predavanje, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)
CROSBI ID: 199568 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
Speaker Recognition Method Combining FFT, Wavelet Functions And Neural Networks
Autori
Grubeša, Sanja ; Grubeša, Tomislav ; Domitrović, Hrvoje
Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni
Izvornik
The Proceedings of the AES 26th International Conference
/ Pritts, Roy - Denver (CO), 2005, 47-50
Skup
AES 26th International Conference, Audio Forensics in the Digital Age
Mjesto i datum
Denver (CO), Sjedinjene Američke Države, 07.07.2005. - 09.07.2005
Vrsta sudjelovanja
Predavanje
Vrsta recenzije
Međunarodna recenzija
Ključne riječi
speaker recognition; neural networks; wavelet functions
Sažetak
The method of speaker recognition based on wavelet functions and neural networks is presented in this paper. The wavelet functions are used to obtain the approximation function and the details of the speaker’ s averaged spectrum in order to extract speaker’ s voice characteristics from the frequency spectrum. The approximation function and the details are then used as input data for decision-making neural networks. In this recognition process, not only the decision on the speaker’ s identity is made, but also the probability that the decision is correct can be provided.
Izvorni jezik
Engleski
Znanstvena područja
Elektrotehnika
POVEZANOST RADA
Ustanove:
Fakultet elektrotehnike i računarstva, Zagreb
