Results (
Thai) 1:
[Copy]Copied!
The results obtained confirm the theory that the addition of the visual modalityincreases the accuracy of ASR systems. It was observed that, while both the audioand visual recognition systems may fail to recognise a particular phoneme, thecombined Audio-Visual recognition is more likely to succeed. This observation isattributed to the fact the Audio-Visual integration scheme maximises the outputprobabilities of both modalities. The synergy criterion has been met for all methodsimplemented on the AVASR platform.
Being translated, please wait..
