A new approach of speaker recognition which combined the advantages of VQ and GMM is presented.

• 摘要提出了一种新的说话人识别方法。