When combined with acoustical speech information, visual speech information (lip movement) significantly improves Automatic Speech Recognition (ASR) in acoustically noisy environments. Previous research has demonstrated that visual modality is a viable tool for identifying speech. However, the visual information has yet to become utilized in mainstream ASR systems due to the difficulty in accurately tracking lips in real-world conditions. This paper presents our current progress in tracking face and lips in visually challenging environments. Findings suggest the mean shift algorithm performs poorly for small regions, in this case the lips, but it achieves near 80% accuracy for facial tracking.

Date Published: 19 January 2009
PDF: 11 pages
Proc. SPIE 7257, Visual Communications and Image Processing 2009, 72571Y (19 January 2009); doi: 10.1117/12.817092

Published in SPIE Proceedings Vol. 7257:
Visual Communications and Image Processing 2009
Majid Rabbani; Robert L. Stevenson, Editor(s)