SENTENCE LIPREADING USING HIDDEN MARKOV MODEL WITH INTEGRATED GRAMMAR

Article Properties
Abstract
Cite
YU, KEREN, et al. “SENTENCE LIPREADING USING HIDDEN MARKOV MODEL WITH INTEGRATED GRAMMAR”. International Journal of Pattern Recognition and Artificial Intelligence, vol. 15, no. 01, 2001, pp. 161-76, https://doi.org/10.1142/s0218001401000770.
YU, K., JIANG, X., & BUNKE, H. (2001). SENTENCE LIPREADING USING HIDDEN MARKOV MODEL WITH INTEGRATED GRAMMAR. International Journal of Pattern Recognition and Artificial Intelligence, 15(01), 161-176. https://doi.org/10.1142/s0218001401000770
YU K, JIANG X, BUNKE H. SENTENCE LIPREADING USING HIDDEN MARKOV MODEL WITH INTEGRATED GRAMMAR. International Journal of Pattern Recognition and Artificial Intelligence. 2001;15(01):161-76.
Journal Categories
Science
Mathematics
Instruments and machines
Electronic computers
Computer science
Technology
Electrical engineering
Electronics
Nuclear engineering
Electronics
Technology
Mechanical engineering and machinery
Description

Can machines learn to lipread entire sentences? This paper presents a systematic approach to sentence lipreading, leveraging a hidden Markov model (HMM) integrated with grammar. It defines a vocabulary of elementary words and a grammar that generates legal sentences. The lipreading approach combines the grammar with HMMs to recognize sequences of words forming valid sentences. The methodology was tested through two experiments: one involving e-mail commands, and another focused on recognizing English integer numbers up to one million. The experiments showed promising results, demonstrating the feasibility of the approach for complex lipreading tasks. This research has implications for the development of advanced human-computer interfaces, assistive technologies for individuals with hearing impairments, and security systems. The approach advances the field of **artificial intelligence** and **communication**, demonstrating the potential of combining linguistic knowledge with statistical models for improved lipreading accuracy.

Published in the International Journal of Pattern Recognition and Artificial Intelligence, this paper is well-suited to the journal's focus on advanced pattern recognition techniques and their applications. The study's use of hidden Markov models and integrated grammar aligns with the journal's emphasis on innovative AI approaches to complex recognition problems.

Refrences