Significance of formants from difference spectrum for speaker identification
نویسندگان
چکیده
In this paper, we describe a prototype speaker identification system using auto-associative neural network (AANN) and formant features. Our experiments demonstrate that formants extracted from difference spectrum perform significantly better than formants extracted from normal spectrum for the task of speaker identification. We also demonstrate that formants from difference spectrum provide comparable speaker identification performance with that of features such as weighted linear predictive Cepstral coefficients and Mel-Frequency Cepstral coefficients. Finally, we combine the results of formant based system and linear predictive Cepstral coefficients based system to achieve 100% identification performance.
منابع مشابه
Wavelet Formants Speaker Identification Based System via Neural Network
In this paper Discrete wavelet Transform with logarithmic Power Spectrum Density (PSD) are combined for speaker formants extraction, to be used as evident classification features. For classification, Feed Forward Back Propagation Neural Network FFBNN method is proposed. The Discrete Wavelet formants Neural Network DWFNNT system works with excellent capability of features tracking even with 0dB ...
متن کاملFormant-broadened CMS using peak-picking in LOG spectrum
In this paper, we propose a method to remove the residual speech effects of the channel cepstrum for speaker recognition in the Cepstral Mean Subtraction framework. The proposed Formant-Broadened CMS(FBCMS) is based on the facts that the formants can be found easily in log spectrum which is transformed from the cepstrum and the formants correspond to the dominant poles of all-pole model which i...
متن کاملThe Use of Group Delay Features of Linear Prediction Model for Speaker Recognition
New text independent speaker identification method is presented. Phase spectrum of allpole linear prediction (LP) model is used to derive the speech features. The features are represented by pairs of numbers that are calculated from group delay extremums of LP model spectrum. The first component of the pair is an argument of maximum of group delay of all pole LP model spectrum and the second is...
متن کاملChannel Compensation for Forensic Speaker Identification Using Inverse Processing
Typically, speaker identification examination requires two audio recordings: a voice sample and a questionable recording. The questionable one is in most of the cases the intercepted or recorded phone call. As mobile phones became the most popular way of communication, the largest number of questionable recordings comes from GSM channels. They use special algorithms and devices to transmit the ...
متن کاملThe prototype model in speaker identification
Little is known on the perceptual processes of speaker identification and its relations to the acoustic features of the speaker’s voice. A study of speaker perception and identification by psycho-acoustic experiments was carried out. Statistical analysis of the results suggests that the prototype model is appropriate for explaining the process of speaker identification. It has been also found t...
متن کامل