• Speech Sciences
• /
• v.12 no.4
• /
• pp.31-42
• /
• 2005

Recently, Gaussian Mixture Model (GMM), a special form of CHMM, has been applied to speaker identification and it has proved that performance of GMM is better than CHMM. Therefore, in this paper the speaker models based on GMM and a new GMM using the modified EM algorithm are introduced and evaluated for text-independent speaker identification. Various experiments were performed to evaluate identification performance of two algorithms. As a result of the experiments, the GMM speaker model attained 94.6% identification accuracy using 40 seconds of training data and 32 mixtures and 97.8% accuracy using 80 seconds of training data and 64 mixtures. On the other hand, the new GMM speaker model achieved 95.0% identification accuracy using 40 seconds of training data and 32 mixtures and 98.2% accuracy using 80 seconds of training data and 64 mixtures. It shows that the new GMM speaker identification performance is better than the GMM speaker identification performance.

• Proceedings of the KSPS conference
• /
• 2003.10a
• /
• pp.185-188
• /
• 2003

In this paper, we construct phonetic GMM for text-independent speaker identification system. The basic idea is to combine of the advantages of baseline GMM and HMM. GMM is more proper for text-independent speaker identification system. In text-dependent system, HMM do work better. Phonetic GMM represents more sophistgate text-dependent speaker model based on text-independent speaker model. In speaker identification system, phonetic GMM using HMM-based speaker-independent phoneme recognition results in better performance than baseline GMM. In addition to the method, N-best recognition algorithm used to decrease the computation complexity and to be applicable to new speakers.

• The Journal of the Acoustical Society of Korea
• /
• v.29 no.3
• /
• pp.216-221
• /
• 2010

In this paper, we propose speaker verification method using Support Vector Machine (SVM) kernel with Gaussian Mixture Model (GMM)-supervector based on the Mahalanobis distance. The proposed GMM-supervector SVM kernel method is combined GMM with SVM. The GMM-supervectors are generated by GMM parameters of speaker and other speaker utterances. A speaker verification threshold of GMM-supervectors is decided by SVM kernel based on Mahalanobis distance to improve speaker verification accuracy. The experimental results for text-independent speaker verification using 20 speakers demonstrates the performance of the proposed method compared to GMM, SVM, GMM-supervector SVM kernel based on Kullback-Leibler (KL) divergence, and GMM-supervector SVM kernel based on Bhattacharyya distance.

• Speech Sciences
• /
• v.10 no.2
• /
• pp.35-43
• /
• 2003

General speaker verification systems improve their recognition performances by normalizing log likelihood ratio, using a speaker model and its background speaker model that are required to be verified. So these systems rely heavily on the availability of much speaker independent databases for background speaker model design. This constraint, however, may be a burden in practical and portable devices such as palm-top computers or wireless handsets which place a premium on computations and memory. In this paper, new approach for the GMM-based background model design used in portable speaker verification system is presented when the enrollment data is available. This approach is to modify three parameters of GMM speaker model such as mixture weights, means and covariances along with reduced mixture order. According to the experiment on a 20 speaker population from YOHO database, we found that this method had a promise of effective use in a portable speaker verification system.

• Speech Sciences
• /
• v.11 no.2
• /
• pp.237-245
• /
• 2004

In general, Gaussian mixture model(GMM) is used to estimate the speaker model for speaker identification. The parameter estimates of the GMM are obtained by using the expectation-maximization (EM) algorithm for the maximum likelihood(ML) estimation. However, if the number of mixtures isn't defined well in the GMM, those parameters are obtained inappropriately. The problem to find the number of components is significant to estimate the optimal parameter in mixture model. In this paper, to estimate the optimal number of mixtures, we propose the method that starts from the sufficient mixtures, after, the number is reduced by investigating the mutual information between mixtures for GMM. In result, we can estimate the optimal number of mixtures. The effectiveness of the proposed method is shown by the experiment using artificial data. Also, we performed the speaker identification applying the proposed method comparing with other approaches.

• MALSORI
• /
• no.45
• /
• pp.93-105
• /
• 2003

In this paper, we present a single framework based on GMM for speaker recognition. The proposed framework can simultaneously minimize environmental variations on mismatched conditions and adapt the bias free and speaker-dependent characteristics of claimant utterances to the background GMM to create a speaker model. We compare the closed-set speaker identification for conventional method and the proposed method both on TIMIT and NTIMIT. In the several sets of experiments we show the improved recognition rates on a simulated channel and a telephone channel condition by 7.2% and 27.4% respectively.

• The KIPS Transactions:PartB
• /
• v.12B no.4 s.100
• /
• pp.437-442
• /
• 2005

For the purpose of speaker normalization in speaker independent speech recognition systems, experiments are conducted on a method based on Gaussian mixture model(GMM). The method, which is an improvement of the previous study based on vector quantizer, consists of modeling the probability distribution of canonical feature vectors by a GMM with an appropriate number of clusters, and of estimating the warp factor of a test speaker by making use of the obtained probabilistic model. The purpose of this study is twofold: improving the existing ML based methods, and comparing the performance of what is called 'soft decision' method with that of the previous study based on vector quantizer. The effectiveness of the proposed method is investigated by recognition experiments on the TIMIT corpus. The experimental results showed that a little improvement could be obtained tv adjusting the number of clusters in GMM appropriately.

• Speech Sciences
• /
• v.10 no.4
• /
• pp.149-157
• /
• 2003

In this paper, we proposed the principal component analysis (PCA) fuzzy mixture model for speaker identification. A PCA fuzzy mixture model is derived from the combination of the PCA and the fuzzy version of mixture model with diagonal covariance matrices. In this method, the feature vectors are first transformed by each speaker's PCA transformation matrix to reduce the correlation among the elements. Then, the fuzzy mixture model for speaker is obtained from these transformed feature vectors with reduced dimensions. The orthogonal Gaussian Mixture Model (GMM) can be derived as a special case of PCA fuzzy mixture model. In our experiments, with having the number of mixtures equal, the proposed method requires less training time and less storage as well as shows better speaker identification rate compared to the conventional GMM. Also, the proposed one shows equal or better identification performance than the orthogonal GMM does.

• MALSORI
• /
• no.47
• /
• pp.121-129
• /
• 2003

This paper describes the use of pitch information for speaker identification. The recognition system is a GMM based one with 4 connected Korean digits speech database. The mean of the pitch period in voiced sections of speech are shown to be ,useful at discriminating between speakers. Utilizing this feature with Gaussian mixture model in the speaker identification system gave a marked improvement, maximum 6% improvement comparing to the baseline Gaussian mixture model.

• Journal of Information Processing Systems
• /
• v.14 no.4
• /
• pp.807-823
• /
• 2018

Speaker verification system performance depends on the utterance of each speaker. To verify the speaker, important information has to be captured from the utterance. Nowadays under the constraints of limited data, speaker verification has become a challenging task. The testing and training data are in terms of few seconds in limited data. The feature vectors extracted from single frame size and rate (SFSR) analysis is not sufficient for training and testing speakers in speaker verification. This leads to poor speaker modeling during training and may not provide good decision during testing. The problem is to be resolved by increasing feature vectors of training and testing data to the same duration. For that we are using multiple frame size (MFS), multiple frame rate (MFR), and multiple frame size and rate (MFSR) analysis techniques for speaker verification under limited data condition. These analysis techniques relatively extract more feature vector during training and testing and develop improved modeling and testing for limited data. To demonstrate this we have used mel-frequency cepstral coefficients (MFCC) and linear prediction cepstral coefficients (LPCC) as feature. Gaussian mixture model (GMM) and GMM-universal background model (GMM-UBM) are used for modeling the speaker. The database used is NIST-2003. The experimental results indicate that, improved performance of MFS, MFR, and MFSR analysis radically better compared with SFSR analysis. The experimental results show that LPCC based MFSR analysis perform better compared to other analysis techniques and feature extraction techniques.