Description:
This revised scholarly work on voice recognition technology outlines cutting-edge research in this exciting area of computer science. The book begins with a readable historical introduction to speech synthesis, speech recognition, and speaker classification. (According to the authors, Alexander Graham Bell was actually working on the problem of speech synthesis when he invented the telephone.) Once the authors move on to their specialty, this text gets increasingly advanced, discussing the multitude of formulas and algorithms used in speech recognition. The writers clearly explain two important areas of recognizing speech, Hidden Markov Models (HMMs) and neural nets as applied to speech. The latter part of the book presents the authors' own research in the field, which involves using the technology to identify characteristics of voices, enabling law enforcement to catch criminals who commit crimes over phone lines. (One of their principal advances, "rehumanizing filter techniques" enables a voice to be identified in even noisy environments.) Though the authors certainly would be the first to admit that voice recognition has a long way to go before it becomes commonplace, they see a strong future for this technology in law and security.
| 
