Speaker recognition using neural network
Speaker recognizer may be employed as part of a security system requiring user authentication. Mostly, years by years there have been several techniques being developed to achieve high success rate of accuracy in the identification and verification of individuals for authentication in security syste...
| Main Author: | |
|---|---|
| Format: | Undergraduates Project Papers |
| Language: | English |
| Published: |
2012
|
| Subjects: | |
| Online Access: | http://umpir.ump.edu.my/id/eprint/7736/ http://umpir.ump.edu.my/id/eprint/7736/ http://umpir.ump.edu.my/id/eprint/7736/1/ZUL_RASYIED_BIN_AB._RASAT.PDF |
| Summary: | Speaker recognizer may be employed as part of a security system requiring user authentication. Mostly, years by years there have been several techniques being developed to achieve high success rate of accuracy in the identification and verification of individuals for authentication in security systems. A major application area of such systems would be providing security for telephone-mediated transaction systems where some form of anatomical or "biometric" identification is desirable. In daily usage, it maybe can be used in car environment or building to voice control non-critical operation such as open the gate to ensure a maximum control to the car or building and enhance the safety. This project is emphasizes on speaker recognizer system which is automatically verifying or recognizing the identity of the speakers based on voice unique characteristics. This system will be focusing on textdependant method and in open-set situation. The feature extraction is done by Mel Frequency Cepstral Coefficients (MFCC). The feature for the speaker who has to be identified are extracted and compared with the stored templates by using Backpropagation Algorithm. After that, the trained network corresponds to the output meanwhile the input is the extracted features of the speaker are identified. The best match of recognize voice is found the speaker identity after the network has done its weight adjustment. For the decision-making purpose, Artificial Neural Network (ANN) method will be used. This project will be done by applying Neural Network Toolbox in MATLAB software. Lastly, the system should be able to automatically accept or reject the voice of different person based on voice unique characteristic. |
|---|