Abstract
Paper presents visual features for speech recognition. Visual speech recognition can be applied as support for acoustical recognition process or as stand-alone speech recognition approach in the case of absence of acoustical data. Geometrical features of speaker‘s lips are presented. Statistical analysis of feature values enabled us to select individual feature for each analyzed phone and one combined feature for all phones. The discriminative power was tested using Bayes classification rule. 4th order analysis enabled us to recognize phones with the average rate of 60.3 %, the highest recognition rate of 81.3 % was achieved identifying phone [o]. DOI: http://dx.doi.org/10.5755/j01.eee.19.5.1724
Highlights
Accurate and reliable speech recognition is the aim for all speech applications and systems
The first way to solve these problems is the employment of robust feature system for acoustical signal analysis
Various feature systems and techniques are applied for this purpose – mean subtraction, mel frequency based cepstral analysis, pre-filtering, speaker adaptation
Summary
Accurate and reliable speech recognition is the aim for all speech applications and systems. Various feature systems and techniques are applied for this purpose – mean subtraction, mel frequency based cepstral analysis, pre-filtering, speaker adaptation. Despite these efforts the acoustical analysis still cannot give robust and reliable speech recognition. Bimodal (audio-visual) recognition of speech is natural for humans some linguistic information is carried by speaker’s view. In this case we can talk about visual speech recognition (in opposition to acoustical speech recognition).
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
