Abstract: This paper draws parallels between speech recognition of one speaker on a limited set of words and recognition of tactile sign language. The paper also provides variant of formation of feature vectors in matrix form for both problems. It is suggested to use ellipsoidal and orthogonal compliance distances.

Keywords: speech and gesture recognition, orthogonal projectors, ellipsoidal distance, pseudoinverse, SVD – decomposition.

ACM Classification Keywords: I.2 Artificial Intelligence, I.4 Image Processing and Computer Vision, I.5 Pattern Recognition, G.1.3 Numerical Linear Algebra.

Link:

MATRIX FEATURE VECTORS IN SPEECH AND GESTURE RECOGNITION

Volodymyr Donchenko, Andrew Golik

http://www.foibg.com/ijita/vol20/ijita20-02-p06.pdf