Finding phonemes: improving machine lip-reading
This work addresses the challenge of improving word recognition accuracy in machine lip-reading systems, which is an incremental advancement in the field.
The paper tackles the problem of determining optimal class units for machine lip-reading by developing speaker-dependent viseme classes through phoneme confusion mapping, showing that phoneme classifiers often outperform viseme classifiers in word recognition on the LiLIR dataset, with intermediate units yielding even better results.
In machine lip-reading there is continued debate and research around the correct classes to be used for recognition. In this paper we use a structured approach for devising speaker-dependent viseme classes, which enables the creation of a set of phoneme-to-viseme maps where each has a different quantity of visemes ranging from two to 45. Viseme classes are based upon the mapping of articulated phonemes, which have been confused during phoneme recognition, into viseme groups. Using these maps, with the LiLIR dataset, we show the effect of changing the viseme map size in speaker-dependent machine lip-reading, measured by word recognition correctness and so demonstrate that word recognition with phoneme classifiers is not just possible, but often better than word recognition with viseme classifiers. Furthermore, there are intermediate units between visemes and phonemes which are better still.