Page 1
00100			The Amanuensis Speech Recognition System
AMANUENSIS
Page 1
00500				James L.Hieronymus
HIERONYMUS
Page 1
00600				  Neil J. Miller
NEIL
Page 1
00700				 Arthur L. Samuel
ARTHUR
SAMUEL
Page 1
00800	The Amanuensis speech recognition system  under  development  at  the
AMANUENSIS
Page 1
01100		1)  A  new acoustic segmenter (2) is used to extract prosodic
ACOUSTIC
PROSODIC
Page 1
01200	features from the acoustic input and to isolate regions for  especial
ACOUSTIC
Page 1
01500	synchronously using a new glottal pulse locator (3).
GLOTTAL
LOCATOR
Page 1
01800		3) Speaker normalization  is  done,  partly  by  formula  and
NORMALIZATION
Page 1
02100		4) Greater use is made of the redundancy of speech to improve
REDUNDANCY
Page 1
02700		6)  A  bootstrapping  technique  is  under study which should
BOOTSTRAPPING
Page 1
02800	greatly reduce the amount of hand segmentation necessary  to  provide
SEGMENTATION
Page 1
02900	the anotated training material.
ANOTATED
Page 1
03000		7) Several possible output streams of phonemes  are  produced
PHONEMES
Page 1
03200	individual phonemes, so that it should not be necessary  ever  to  go
PHONEMES
Page 1
03300	back  to  the original acoustic input data to resolve ambiguities and
ACOUSTIC