Glottal modeling and closed-phase analysis for speaker recognition
نویسندگان
چکیده
This paper concerns the application of glottal models and closed-phase analysis to the problem of speaker recognition. A glottal model based on one originally proposed by Fujisaki and Ljungqvist was used in conjunction with closed-phase analysis to yield features for a speaker recognition system used in the NIST 2003 Speaker Recognition Evaluation. Scores from the system based on the glottal model features were combined with scores from a system using formant center frequencies and bandwidths and F0 (FMBWF0), yielding significant improvement over the FMBWF0 system alone. The combination of the glottal model and FMBWF0 scores was in turn combined with the scores from a standard MFCC system to yield improvement beyond that of the MFCC system alone.
منابع مشابه
Glottal-based analysis of the lombard effect
The Lombard effect refers to the speech changes due to the immersion of the speaker in a noisy environment. Among these changes, studies have already reported acoustic modifications mainly related to the vocal tract behaviour. In a complementary way, this paper investigates the variation of the glottal flow in Lombard speech. For this, the glottal flow is estimated by a closed-phase analysis an...
متن کاملModeling of the glottal flow derivative waveform with application to speaker identification
Speech production has long been viewed as a linear filtering process, as described by Fant in the late 1950's [10]. The vocal tract, which acts as the filter, is the primary focus of most speech work. This thesis develops a method for estimating the source of speech, the glottal flow derivative. Models are proposed for the coarse and fine structure of the glottal flow derivative, accounting for...
متن کاملVoice source cepstrum processing for speaker identification
Voice source analysis and modelling has played a key role in important speech applications such as speech recognition, speech synthesis and speaker recognition. This work presents a robust algorithm for glottal closure detection and a novel set of voice source features for speaker recognition. In the rst part of the dissertation the DYPSA algorithm is developed for detecting glottal closure ins...
متن کاملSpeaker Identification Using Glottal-Source Waveforms and Support-Vector-Machine Modelling
Speaker identification experiments are performed with novel features representative of the glottal source waveform. These are derived from closed-phase analysis and inverse filtering. Source waveforms are segmented into two consecutive periods and normalised in prosody, forming so called source-frame feature vectors. Support-vector-machines are used to construct speaker discriminative hyperplan...
متن کاملSpeaker Verification Using the Shape of the Glottal Excitation Function for Vowels
This paper seeks to establish a baseline for the potential contribution of the shape of the glottal source waveform to speaker recognition. A text-dependent speaker verification experiment was performed with 4 monosyllabic words spoken repeatedly by the 16 speakers of the TI46 speech data corpus. A single fundamental period was automatically extracted from each vowel centre and inverse-filtered...
متن کامل