Pass-Phrase based Speaker Identification

The problem of speaker identification is an area with many different applications. The most practical use can be found in applications dealing with security, surveillance, and automatic transcription in a multi-speaker environment. Speaker identification is a difficult task and the task has several different approaches. The state of the art for speaker identification techniques include Dynamic Time Warped (DTW) template matching, Hidden Markov Modeling (HMM), and codebook schemes based on Vector Quantization (VQ). This paper emphasizes on text dependent speaker identification, which deals with detecting a particular speaker from a known population. The system reads the speech utterance. System identifies the user by comparing the codebook of speech utterance with those of the stored in the database and lists, which contain the most likely speakers, could have given that speech utterance. The vector quantization approach will be proposed, due to ease of implementation and high accuracy.