Fast methods for code search in CELP

The code excited linear prediction (CELP) technique has the potential for producing high quality synthetic speech at bit rates as low as 4.8 kb/s. Three fast search methods for selecting an optimal excitation sequence from a code book of stochastic vectors are described. The key idea is to inverse-f...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Ahmed, M.E. (author)
مؤلفون آخرون: Al-Suwaiyel, M.I. (author), unknown (author)
التنسيق: article
منشور في: 1993
الموضوعات:
الوصول للمادة أونلاين:https://eprints.kfupm.edu.sa/id/eprint/14491/1/14491_1.pdf
https://eprints.kfupm.edu.sa/id/eprint/14491/2/14491_2.doc
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1864513393812045824
author Ahmed, M.E.
author2 Al-Suwaiyel, M.I.
unknown
author2_role author
author
author_facet Ahmed, M.E.
Al-Suwaiyel, M.I.
unknown
author_role author
dc.creator.none.fl_str_mv Ahmed, M.E.
Al-Suwaiyel, M.I.
unknown
dc.date.none.fl_str_mv 1993-07
2020
dc.format.none.fl_str_mv application/pdf
application/msword
dc.identifier.none.fl_str_mv https://eprints.kfupm.edu.sa/id/eprint/14491/1/14491_1.pdf
https://eprints.kfupm.edu.sa/id/eprint/14491/2/14491_2.doc
(1993) Fast methods for code search in CELP. Speech and Audio Processing, IEEE Transactions on, 1.
dc.language.none.fl_str_mv en
en
dc.publisher.none.fl_str_mv IEEE
dc.relation.none.fl_str_mv https://eprints.kfupm.edu.sa/id/eprint/14491/
dc.rights.*.fl_str_mv info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Computer
dc.title.none.fl_str_mv Fast methods for code search in CELP
dc.type.none.fl_str_mv Article
PeerReviewed
info:eu-repo/semantics/publishedVersion
info:eu-repo/semantics/article
description The code excited linear prediction (CELP) technique has the potential for producing high quality synthetic speech at bit rates as low as 4.8 kb/s. Three fast search methods for selecting an optimal excitation sequence from a code book of stochastic vectors are described. The key idea is to inverse-filter the actual speech with the formant and pitch filters to produce a residual error sequence (RES). The residual error is used to identify a neighborhood or a subset of codes for further processing. The first method, called dynamic nearest neighborhood (DNN), attempts to dynamically construct a neighborhood of the K codes of maximum correlation with the residual error. The second method, called nearest fixed neighborhood (NFN), clusters the code book into a fixed number of cells, and code search is performed on the codes of the cell nearest to the RES. The two methods achieve a reduction in the search procedure by a factor of 8-20 times. The third method combines the advantages of the first two methods to attain a reduction from 40 to 50 times. The performance of these techniques and some of their ramifications are addressed
eu_rights_str_mv openAccess
format article
id KFUPM_7d201792206e1a0acc0cff724a7ae6b1
identifier_str_mv (1993) Fast methods for code search in CELP. Speech and Audio Processing, IEEE Transactions on, 1.
language_invalid_str_mv en
network_acronym_str KFUPM
network_name_str King Fahd University of Petroleum and Minerals
oai_identifier_str oai::14491
publishDate 1993
publisher.none.fl_str_mv IEEE
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
spelling Fast methods for code search in CELPAhmed, M.E.Al-Suwaiyel, M.I.unknownComputerThe code excited linear prediction (CELP) technique has the potential for producing high quality synthetic speech at bit rates as low as 4.8 kb/s. Three fast search methods for selecting an optimal excitation sequence from a code book of stochastic vectors are described. The key idea is to inverse-filter the actual speech with the formant and pitch filters to produce a residual error sequence (RES). The residual error is used to identify a neighborhood or a subset of codes for further processing. The first method, called dynamic nearest neighborhood (DNN), attempts to dynamically construct a neighborhood of the K codes of maximum correlation with the residual error. The second method, called nearest fixed neighborhood (NFN), clusters the code book into a fixed number of cells, and code search is performed on the codes of the cell nearest to the RES. The two methods achieve a reduction in the search procedure by a factor of 8-20 times. The third method combines the advantages of the first two methods to attain a reduction from 40 to 50 times. The performance of these techniques and some of their ramifications are addressedIEEE1993-072020ArticlePeerReviewedinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articleapplication/pdfapplication/mswordhttps://eprints.kfupm.edu.sa/id/eprint/14491/1/14491_1.pdfhttps://eprints.kfupm.edu.sa/id/eprint/14491/2/14491_2.doc (1993) Fast methods for code search in CELP. Speech and Audio Processing, IEEE Transactions on, 1. enenhttps://eprints.kfupm.edu.sa/id/eprint/14491/info:eu-repo/semantics/openAccessoai::144912019-11-01T14:06:04Z
spellingShingle Fast methods for code search in CELP
Ahmed, M.E.
Computer
status_str publishedVersion
title Fast methods for code search in CELP
title_full Fast methods for code search in CELP
title_fullStr Fast methods for code search in CELP
title_full_unstemmed Fast methods for code search in CELP
title_short Fast methods for code search in CELP
title_sort Fast methods for code search in CELP
topic Computer
url https://eprints.kfupm.edu.sa/id/eprint/14491/1/14491_1.pdf
https://eprints.kfupm.edu.sa/id/eprint/14491/2/14491_2.doc