Speaker Recognition Technology
Tomoko Matsui, Sadaoki Furui, Sadaoki Furui · Tokyo Tech Research Repository (Tokyo Institute of Technology) · 2009
Speaker recognition is the process of automatically recognizing who is speaking. This paper introduces three speaker recognition methods; a text-dependent method which uses predetermined keywords, a text-independent method which does not rely on a specific text being spoken, and a textprompted method in which the system prompts each user with a new key sentence every time the system is used. Recently, AT & T and TI (with US Sprint) started field tests and real application of speaker recognition technology, and the Sprint's Voice Phone Card has already been used by many customers. However, there still remain many technical problems to be solved. Among them, this paper introduces several issues that have recently been investigated at NTT Laboratories