A Comparison of Language Processing Techniques for a Constrained Speech Translation System

M. Lincoln; S. J. Cox

doi:10.1109/ICASSP.2003.1198855

A Comparison of Language Processing Techniques for a Constrained Speech Translation System

M. Lincoln, S. J. Cox

Research output: Contribution to conference › Paper

2 Citations (Scopus)

Abstract

A system designed to allow Post Office counter clerks to communicate with deaf customers by translating speech into sign language is described. The system uses approximately 370 pre-stored phrases which may be signed to the customer using a specially designed avatar. The clerk is unable to memorise this number of phrases and therefore the system attempts to map from their input speech to the semantically equivalent pre-stored phrase. We describe a number of language processing techniques developed to perform the mapping, and give results obtained using alternative formulations of the phrases from a number of speakers. We then give results for recognised speech input and show how misrecognitions effect the mapping system. Best performance is obtained using a mapping system based on an entropy weighted, vector based distance measure between the test phrase and each of the signed phrases.

Original language	English
Pages	612-615
Number of pages	4
DOIs	https://doi.org/10.1109/ICASSP.2003.1198855
Publication status	Published - 2003
Event	IEEE Conference on Acoustics, Speech and Signal Processing (ICASSP '03) - Hong Kong Duration: 6 Apr 2003 → 10 Apr 2003

Conference

Conference	IEEE Conference on Acoustics, Speech and Signal Processing (ICASSP '03)
City	Hong Kong
Period	6/04/03 → 10/04/03

Access to Document

10.1109/ICASSP.2003.1198855

Cite this

@conference{5672617641624a0e80ad4ec44f3ef5a1,

title = "A Comparison of Language Processing Techniques for a Constrained Speech Translation System",

abstract = "A system designed to allow Post Office counter clerks to communicate with deaf customers by translating speech into sign language is described. The system uses approximately 370 pre-stored phrases which may be signed to the customer using a specially designed avatar. The clerk is unable to memorise this number of phrases and therefore the system attempts to map from their input speech to the semantically equivalent pre-stored phrase. We describe a number of language processing techniques developed to perform the mapping, and give results obtained using alternative formulations of the phrases from a number of speakers. We then give results for recognised speech input and show how misrecognitions effect the mapping system. Best performance is obtained using a mapping system based on an entropy weighted, vector based distance measure between the test phrase and each of the signed phrases.",

author = "M. Lincoln and Cox, {S. J.}",

year = "2003",

doi = "10.1109/ICASSP.2003.1198855",

language = "English",

pages = "612--615",

note = "IEEE Conference on Acoustics, Speech and Signal Processing (ICASSP '03) ; Conference date: 06-04-2003 Through 10-04-2003",

}

TY - CONF

T1 - A Comparison of Language Processing Techniques for a Constrained Speech Translation System

AU - Lincoln, M.

AU - Cox, S. J.

PY - 2003

Y1 - 2003

N2 - A system designed to allow Post Office counter clerks to communicate with deaf customers by translating speech into sign language is described. The system uses approximately 370 pre-stored phrases which may be signed to the customer using a specially designed avatar. The clerk is unable to memorise this number of phrases and therefore the system attempts to map from their input speech to the semantically equivalent pre-stored phrase. We describe a number of language processing techniques developed to perform the mapping, and give results obtained using alternative formulations of the phrases from a number of speakers. We then give results for recognised speech input and show how misrecognitions effect the mapping system. Best performance is obtained using a mapping system based on an entropy weighted, vector based distance measure between the test phrase and each of the signed phrases.

AB - A system designed to allow Post Office counter clerks to communicate with deaf customers by translating speech into sign language is described. The system uses approximately 370 pre-stored phrases which may be signed to the customer using a specially designed avatar. The clerk is unable to memorise this number of phrases and therefore the system attempts to map from their input speech to the semantically equivalent pre-stored phrase. We describe a number of language processing techniques developed to perform the mapping, and give results obtained using alternative formulations of the phrases from a number of speakers. We then give results for recognised speech input and show how misrecognitions effect the mapping system. Best performance is obtained using a mapping system based on an entropy weighted, vector based distance measure between the test phrase and each of the signed phrases.

U2 - 10.1109/ICASSP.2003.1198855

DO - 10.1109/ICASSP.2003.1198855

M3 - Paper

SP - 612

EP - 615

T2 - IEEE Conference on Acoustics, Speech and Signal Processing (ICASSP '03)

Y2 - 6 April 2003 through 10 April 2003

ER -