Speaker Verification Using Complementary Information from Vocal Source and Vocal Tract
نویسندگان
چکیده
This paper describes a speaker verification system which uses two complementary acoustic features: Mel-frequency cepstral coefficients (MFCC) and wavelet octave coefficients of residues (WOCOR). While MFCC characterizes mainly the spectral envelope, or the formant structure of the vocal tract system, WOCOR aims at representing the spectro-temporal characteristics of the vocal source excitation. Speaker verification experiments carried out on the ISCSLP 2006 SRE database demonstrate the complementary contributions of MFCC and WOCOR to speaker verification. Particularly, WOCOR performs even better than MFCC in single channel speaker verification task. Combining MFCC and WOCOR achieves higher performance than using MFCC only in both single and cross channel speaker verification tasks.
منابع مشابه
Comparative Analysis of Discrimination Power of the Vocal Source and Vocal Tract Features for Speaker Verification
The paper comparatively analyzes the speaker discrimination power of the vocal source and vocal tract related features and present a speaker verification system optimally utilizing the source and tract related speaker specific information. A pitchsynchronous wavelet transform is adopted to capture the speaker specific information from the vocal source signal, particularly the Linear Prediction ...
متن کاملIntegrating Complementary Features from Vocal Source and Vocal Tract for Speaker Identification
This paper describes a speaker identification system that uses complementary acoustic features derived from the vocal source excitation and the vocal tract system. Conventional speaker recognition systems typically adopt the cepstral coefficients, e.g., Mel-frequency cepstral coefficients (MFCC) and linear predictive cepstral coefficients (LPCC), as the representative features. The cepstral fea...
متن کاملTime –Frequency Representation of Vocal Source Signal for Speaker Verification
We propose an effective feature extraction technique for obtaining essential time-frequency information from the linear prediction (LP) residual signal, which are closely related to the glottal vibration of individual speaker. With pitch synchronous analysis, wavelet transform is applied to every two pitch cycles of the LP residual signal to generate a new feature vector, called Wavelet Based F...
متن کاملLong term measures of the resonating vocal tract: establishing correlation and complementarity
Underlying much of the research in forensic voice comparison (FVC) is the assumption that the vocal tract is a useful biometric for speaker discrimination and that individual differences in its anatomy and physiology will be reflected as speech resonances that are recoverable from its output. There are many ways in which the output of the tract may be observed and analysed, different methods de...
متن کاملSpeaker Identification by Combining Various Vocal Tract and Vocal Source Features
Previously, we proposed a speaker recognition system using a combination of MFCC-based vocal tract feature and phase information which includes rich vocal source information. In this paper, we investigate the efficiency of combination of various vocal tract features (MFCC and LPCC) and vocal source features (phase and LPC residual) for normal-duration and short-duration utterance. The Japanese ...
متن کامل