1976
DOI: 10.1109/tc.1976.1674624
|View full text |Cite
|
Sign up to set email alerts
|

The Hearsay-I Speech Understanding System: An Example of the Recognition Process

Abstract: In l4enrsay-.I ciivcrsc sources of kt~owledpc can be represented as cooj~ernting independcat par:~llel processes which help i n the decodil~g of the utterssces using lllc hypotl~esize-r~ld-test pur;ldigol. , 'the system is discussed by collsiderillg a specific esn~iiple of its ol~eratiotl in the realm of voice chess. ~Gpics: feature extraction and sqnle~lt:ition, the recognition process, speaker-and environment-dcpaldent knowledge. sy~itnctic nnd semantic knowl6dge.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
26
0

Year Published

1979
1979
2015
2015

Publication Types

Select...
5
2
2

Relationship

0
9

Authors

Journals

citations
Cited by 64 publications
(26 citation statements)
references
References 5 publications
0
26
0
Order By: Relevance
“…Within the verbal domain itself, we can distinguish subsystems for lexical, syntactic, semantic, and perhaps pragmatic information . Simulations of human language use have found it convenient to treat these systems as semi-autonomous-partially separate but highly interactive (Nash-Webber, 1975 ;Reddy, Erman, Fennell, & Neely, 1973 ;Winograd, 1972) . We can find other such specialized 374 GEN fNF.R AND LOF TS subsystems ; for example, in modelling human understanding of electronic circuitry, knowledge of procedures and knowledge of enduring facts have been conceived of as separate interactive systems (Brown & Burton, 1975) .…”
Section: Discussionmentioning
confidence: 99%
“…Within the verbal domain itself, we can distinguish subsystems for lexical, syntactic, semantic, and perhaps pragmatic information . Simulations of human language use have found it convenient to treat these systems as semi-autonomous-partially separate but highly interactive (Nash-Webber, 1975 ;Reddy, Erman, Fennell, & Neely, 1973 ;Winograd, 1972) . We can find other such specialized 374 GEN fNF.R AND LOF TS subsystems ; for example, in modelling human understanding of electronic circuitry, knowledge of procedures and knowledge of enduring facts have been conceived of as separate interactive systems (Brown & Burton, 1975) .…”
Section: Discussionmentioning
confidence: 99%
“…One kind of model, which grows out of speech engineering and artificial intelligence, attempts to provide a machine solution to the problem of speech recognition. Examples of this kind of model are HEARSAY (Reddy, Erman, Fennell, and Neely, 1973;Erman & Lesser, 1980), HWIM (Wolf & Woods, 1978), HARPY (Lowerre, 1976), and LAFS/SCRIBER (Klatt, 1980). A second kind of model, growing out of experimental psychology, attempts to account for aspects of psychological data on the perception of speech.…”
Section: Criteria and Constraints On Model Developmentmentioning
confidence: 99%
“…The inspiration for the architecture of TRACE goes back to the HEARSAY Speech understanding system (Reddy et al, 1973;Erman and Lesser, 1980). HEARSAY introduced the notion of a Blackboard, a structure similar to the Trace in the TRACE model.…”
Section: The Trace Modelmentioning
confidence: 99%
“…Similar distinctions have been drawn by others, particularly in computational approaches to visual modeling. One early example was the separation between different knowledge sources and the "blackboard" through which they communicate (Reddy, Erman, Farrell, & Neely, 1973). This idea was extended by McClelland (1986) within the connectionist PDP framework to allow more than one word to be identified at a time, using what he called "programmable blackboards."…”
Section: (3) Object Tokensmentioning
confidence: 99%