Indian Institute of Information Technology and Management Gwalior 24/12/2008

DR. ANUPAM SHUKLADR. RITU TIWARI

HEMANT KUMAR MEENARAHUL KALA

Speaker Identification Using Wavelet Analysis and

ANN

Shukla, Anupam; Tiwari, Ritu; Meena, Hemant Kumar & Kala, Rahul; “Speaker Identification using Wavelet Analysis and Artificial Neural Networks”, proceedings of the National Symposium on Acoustics (NSA) 2008

Indian Institute of Information Technology and Management Gwalior 24/12/2008

1. INTRODUCTION2. TECHNIQUES USED3. PROCEDURE4. RESULTS5. CONCLUSION

Index

Indian Institute of Information Technology and Management Gwalior 24/12/2008

IntroductionIdentification of a person is a very traditional

problem.

Finger print recognition, face recognition, signature recognition are common techniques.

Speaker recognition or Automatic Speaker Identification (ASI) identifies an author based on the words spoken.

We have used wavelet analysis to extract the various features and Artificial Neural Networks to identify the speaker by the extracted features.

Indian Institute of Information Technology and Management Gwalior 24/12/2008

Common Techniques

1. Analysis techniquesFourier AnalysisShort Time Fourier AnalysisWavelet Analysis

2. Artificial Neural Networks

Indian Institute of Information Technology and Management Gwalior 24/12/2008

Analysis Techniques

We have used Wavelet transform to extract characteristics, which is an advancement over Fourier analysis and Short Time Fourier Analysis (STFT).

Indian Institute of Information Technology and Management Gwalior 24/12/2008

Fourier Analysis

Indian Institute of Information Technology and Management Gwalior 24/12/2008

Short Time Fourier Analysis

Indian Institute of Information Technology and Management Gwalior 24/12/2008

Wavelet Analysis

It is a windowing technique with variable-sized regions.

Wavelet analysis allows the use of different time intervals for different type frequency information.

Indian Institute of Information Technology and Management Gwalior 24/12/2008

Wavelet Analysis(Cont..)

Indian Institute of Information Technology and Management Gwalior 24/12/2008

Wavelet Analysis(Cont..)

Capable of revealing aspects of data

Wavelet packet methodSignal decomposition

Indian Institute of Information Technology and Management Gwalior 24/12/2008

Wavelet Packet Analysis

Indian Institute of Information Technology and Management Gwalior 24/12/2008

Artificial Neural Network

Excellent means of machine learning

Reputed training of the system to learn the given data

Testing

Performance

Indian Institute of Information Technology and Management Gwalior 24/12/2008

Procedure

Collection of data sets

Analysis of data sets (feature extraction)

Training of ANN

Testing

Result

Indian Institute of Information Technology and Management Gwalior 24/12/2008

Normalization of Data

Ii=(Vi - Mean(Vij) ) / (Max(Vij) - Mean(Vij) ), for all j

Here Ii is th ith input of the neural network

Vi is the ith feature extracted from Wavelet Analysis

Mean(Vi) is the mean of all Vij found in the training data set

Max(Vi) is the maximum of all Vij found in training data set for all j in data set

Indian Institute of Information Technology and Management Gwalior 24/12/2008

Feature Extracted

Indian Institute of Information Technology and Management Gwalior 24/12/2008

Result

Performance of 97.5%

This clearly shows that the algorithm works well and gives correct results on almost all inputs.

20 speakers and 40 test cases (39 correctly identified)

Indian Institute of Information Technology and Management Gwalior 24/12/2008

Conclusions