Waveform Model of Vowel Perception and Production

Business Announcements
Waveform Communication is prepared to license each aspect of the patent filed on 9/23/11, and establish additional business and academic relationships. The patent protects the Waveform Model of Vowels, developed speech recognition algorithms, improvements to speech in noise, improvements to hearing aids, and talker identification. Pricing will change once the 2011 Acoustic Society of America presentation of 99.2% accuracy across 50 talkers is published.

Michael Stokes - CEO

Email:
waveform.model@yahoo.com

Waveform Communication was named one of the top 50 start-ups in the world in November, 2011 (GEW 50 was established by the Kauffman Foundation).

Paperback - $29.95
eBook - $27
(pricing varies by site)

The Waveform Model of Vowels has been published by
Universal Publishers.
The book can be purchased directly at
Universal Publishers following this link.

The book is also available at the following:

Waveform Model of Vowels
97.7% accuracy across 20+ talkers in the Waveform Model book
99.2% accuracy across the Hillenbrand dataset (1995) - Presented in 2011

Synopsis
The Waveform Model book presents a new model of vowel perception and production derived from visual cues identified in waveform displays. In addition to describing waveform displays of vowels beyond previous descriptions, included in the book are descriptions of experimental evidence supporting near 100% vowel identification accuracy across 20 male talkers using the concepts in the Waveform Model. The book content will be of interest to several academic fields including Cognitive Science, Psychology, Linguistics, Speech and Hearing, Language Acquisition, Neurolinguistics, Phonetics, and areas within Physics and Mathematics. Beyond these academic fields, many speech recognition programs are based on simple statistical programs like Hidden Markov Models that ignore any theoretical basis to speech recognition. The Waveform Model differs from the HMM approaches since it has a theoretical basis rooted in articulation and that has potentially more promise than these simple HMM models that just take overall similarities in waveforms and try to match them to phonemes and words. Furthermore, many of the speech recognition programs use extensive training by a single user (in quiet conditions) in order to attain over 95% accuracy, which is still a relatively poor performance. The Waveform Model can be used across talkers, has been used to achieve 99.2% accuracy across 50 talkers, and has accuracy above reported speech recognition performance (specific to vowels). Multiple applications are being developed from the innovations and success of the Waveform Model.

"It is not possible to look at the waveform of an utterance and say what sounds occurred", (Ladefoged, 1982, A Course in Phonetics, p. 168).

The impossible is now possible.