Speaker recognition based on VQ
Tan Rui-lian · Information technology newsletter · 2010
Speaker recognition is a special mode of speech recognition.The intention does not to recognize speech signals.It expects who is speaking.That means distilling one's character from the original speech signals.Using VQ not only can avoid difficult speech subsection and time warping,but also can reduce data store as a constringent method.Thought the investigation of the speaker recognition,this text designs a system of speaker recognition.When the data of training are small,complex cepstrum can recognize stably.