The ‘state of the art’ in speech recognition is reviewed with particular reference to the kind of problems that are likely to arise in a parcel sorting environment. Speech recognition equipment developed by the authors is described. To justify a speech recognition equipment for parcel sorting it must be shown to increase productivity. Simulations relevant to voice control of parcel sorting have been carried out to try to assess what this improvement might be, and the results are discussed.
Get full access to this article
View all access options for this article.
References
1.
LindgrenN.‘Machine recognition of human language’, IEEE Spectrum19652, 3 (Pt 1), 4 (Pt 2), 5 (Pt 3).
2.
DavisK. H.BiddulphR. S.BalashekS.‘Automatic recognition of spoken digits’, J. acoust. Soc. Am.195222 (6), 639.
3.
SebestynG. S.Decision-making processes in pattern recognition (1st edn) 1962 (Macmillan, New York).
4.
PurtonR. F.‘Speech recognition using auto-correlation analysis’, Paper B8, Conference on Speech Communication and Processing, Boston 1967.
5.
BezdelW.BridleJ. S.‘Speech recognition using zero crossing measurements and sequence information’, Proc. Instn elect. Engrs1969116, 617.
6.
ScarrR. W. A.‘Word-recognition machine’, Proc. Instn elect. Engrs1970117, 203.
7.
MartinT. B.ZadellH. J.CoxR. B.NelsonA. L.‘Applications of automatic speech recognition techniques to bandwidth compression’, RCA Engr196712 (5), 64.
8.
SholtzP. N.BakisR.‘Spoken digit recognizer using vowel consonant segmentation’, J. acoust. Soc. Am.196234, 1.
9.
United States Information Service.‘Voice recognition machines that listen’, Sci. Horiz.1969 (June), No. 105.
