Text Independent Speaker Identification Using Gaussian Mixture Model

1. Text Independent Speaker Identification Using Gaussian Mixture Model Chee-Ming Ting Sh-HussainSallehTian-Swee Tan A. K. Ariff. International Conference on Intelligent and Advanced Systems 2007 Jain-De,Lee

2. OUTLINE • INTRODUCTION • GMM SPEAKER IDENTIFICATION SYSTEM • EXPERIMENTAL EVALUATION • CONCLUSION

3. INTRODUCTION • Speaker recognition is generally divided into two tasks • Speaker Verification(SV) • Speaker Identification(SI) • Speaker model  • Text-dependent(TD) • Text-independent(TI)

4. INTRODUCTION • Many approaches have been proposed for TI speaker recognition • VQ based method • Hidden Markov Models • Gaussian Mixture Model • VQ based method

5. INTRODUCTION • Hidden Markov Models • State Probability • Transition Probability • Classify acoustic events corresponding to HMM states to characterize each speaker in TI task • TI performance is unaffected by discarding transition probabilities in HMM models

6. INTRODUCTION • Gaussian Mixture Model • Corresponds to a single state continuous ergodic HMM • Discarding the transition probabilities in the HMM models • The use of GMM for speaker identity modeling • The Gaussian components represent some general speaker-dependent spectral shapes • The capability of Gaussian mixture to model arbitrary densities

7. GMM SPEAKER IDENTIFICATION SYSTEM • The GMM speaker identification system consists of the following elements • Speech processing • Gaussian mixture model • Parameter estimation • Identification

8. Speech Processing • The Mel-scale frequency cepstral coefficients (MFCC) extraction is used in front-end processing Input Speech Signal Pre-Emphasis Frame Hamming Window FFT Triangular band-pass filter Logarithm DCT Mel-sca1e cepstral feature analysis

9. Gaussian mixture model • The Gaussian model is a weighted linear combination of M uni-model Gaussian component densities • The mixture weight satisfy the constraint that Where is a D-dimensional vector are the component densities wi, i=1,…,Mare the mixture weights

10. Gaussian mixture model • Each component density is a D-variate Gaussian function of the form • The Gaussian mixture density model are denoted as Where is mean vector is covariance matrix

11. Parameter estimation • Conventional GMM training process Input training vector LBG algorithm N EM algorithm Convergence End Y

12. LBG Algorithm Input training vector Overall average N Y m<M End Split D’=D Clustering Cluster’s average  N Y Calculate Distortion   (D-D’)/D< δ

13. EM Algorithm • Speaker model training is to estimate the GMM parameters via maximum likelihood (ML) estimation • Expectation-maximization (EM) algorithm

14. Parameter estimation • This paper proposes an algorithm consists of two steps

15. Parameter estimation • Cluster the training vectors to the mixture component with the highest likelihood • Re-estimate parameters of each component number of vectors classified in cluster i / total number of training vectors sample mean of vectors classified in cluster i. sample covariance matrix of vectors classified in cluster i

16. Identification • The feature is classified to the speaker ,whosemodel likelihood is the highest • The above can be formulated in logarithmic term

17. EXPERIMENTAL EVALUATION • Database and Experiment Conditions • 7 male and 3 female • The same 40 sentences utterances with different text • The average sentences duration is approximately 3.5 s • Performance Comparison between EM and Highest Mixture Likelihood Clustering Training • The number of Gaussian components 16 • 16 dimensional MFCCs • 20 utterances is used for training

18. EXPERIMENTAL EVALUATION • Convergence condition

19. EXPERIMENTAL EVALUATION • The comparison between EM and highest likelihood clustering training on identification rate • 10 sentences were used for training • 25 sentences were used for testing • 4 Gaussian components • 8 iterations

20. EXPERIMENTAL EVALUATION • Effect of Different Number of Gaussian Mixture Components and Amount of Training Data • MFCCs feature dimension is fixed to 12 • 25 sentences is used for testing

21. EXPERIMENTAL EVALUATION • Effect of Feature Set on Performance for Different Number of Gaussian Mixture Components • Combination with first and second order difference coefficients was tested • 10 sentences is used for training • 30 sentences is used for testing

22. CONCLUSION • Comparably to conventional EM training but with less computational time • First order difference coefficients is sufficient to capture the transitional information with reasonable dimensional complexity • The 12 dimensional 16 order GMM and using 5 training sentences achieved 98.4% identification rate