Comparative Study on Feature Extraction and Classifier Techniques for Speaker Recognition
APARNA K. MOHAN, A Sulphikar · Journal of Emerging Technologies and Innovative Research · 2015
In the real world when we hear a voice through a telephone or at an entry way out of sight we identify the person through the sound of their voice. It is based on the expectation that the sound of one’s voice is sufficient for a hearer to recognize the speaker. This is considered to be a challenge for speaker recognition. Voice biometrics use the features of a person’s voice to ascertain their identity like human listeners. Speaker recognition is the best known commercialized form of voice biometrics. Speaker recognition is a process that verify a person’s claim of his/her identity using the features extracted from their voice. The speech signal carries information about the speaker, the message to be conveyed, and language emotion and so on. In this paper we have presented a study on speaker recognition and various feature extraction and classifier techniques. First section gives an introduction to speaker recognition. The next section provide a brief description about the key terms that is used in speaker recognition and then discusses the different feature extraction techniques and classifier techniques for speaker recognition