{"id":"https://openalex.org/W2168331946","doi":"https://doi.org/10.1109/icassp.2006.1660025","title":"Automatic Speech Attribute Transcription (ASAT) - The Front End Processor","display_name":"Automatic Speech Attribute Transcription (ASAT) - The Front End Processor","publication_year":2006,"publication_date":"2006-08-02","ids":{"openalex":"https://openalex.org/W2168331946","doi":"https://doi.org/10.1109/icassp.2006.1660025","mag":"2168331946"},"language":"en","primary_location":{"id":"doi:10.1109/icassp.2006.1660025","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp.2006.1660025","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2006 IEEE International Conference on Acoustics Speed and Signal Processing Proceedings","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101603082","display_name":"Jun Hou","orcid":"https://orcid.org/0000-0001-9437-4035"},"institutions":[{"id":"https://openalex.org/I102322142","display_name":"Rutgers, The State University of New Jersey","ror":"https://ror.org/05vt9qd57","country_code":"US","type":"education","lineage":["https://openalex.org/I102322142"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jun Hou","raw_affiliation_strings":["CAIP Center, Rutgers University, Piscataway, NJ, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"CAIP Center, Rutgers University, Piscataway, NJ, USA","institution_ids":["https://openalex.org/I102322142"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5110218997","display_name":"L. R. Rabiner","orcid":null},"institutions":[{"id":"https://openalex.org/I102322142","display_name":"Rutgers, The State University of New Jersey","ror":"https://ror.org/05vt9qd57","country_code":"US","type":"education","lineage":["https://openalex.org/I102322142"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"L. Rabiner","raw_affiliation_strings":["CAIP Center, Rutgers University, Piscataway, NJ, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"CAIP Center, Rutgers University, Piscataway, NJ, USA","institution_ids":["https://openalex.org/I102322142"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5112247126","display_name":"Sorin Dusan","orcid":null},"institutions":[{"id":"https://openalex.org/I102322142","display_name":"Rutgers, The State University of New Jersey","ror":"https://ror.org/05vt9qd57","country_code":"US","type":"education","lineage":["https://openalex.org/I102322142"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"S. Dusan","raw_affiliation_strings":["CAIP Center, Rutgers University, Piscataway, NJ, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"CAIP Center, Rutgers University, Piscataway, NJ, USA","institution_ids":["https://openalex.org/I102322142"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I102322142"],"apc_list":null,"apc_paid":null,"fwci":3.7294,"has_fulltext":false,"cited_by_count":11,"citation_normalized_percentile":{"value":0.93354423,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":89,"max":94},"biblio":{"volume":"1","issue":null,"first_page":"I","last_page":"333"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9998000264167786,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9998000264167786,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":0.9979000091552734,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11309","display_name":"Music and Audio Processing","score":0.9872000217437744,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7662114500999451},{"id":"https://openalex.org/keywords/front-and-back-ends","display_name":"Front and back ends","score":0.6265080571174622},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5741527676582336},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.562653660774231},{"id":"https://openalex.org/keywords/probabilistic-logic","display_name":"Probabilistic logic","score":0.5477754473686218},{"id":"https://openalex.org/keywords/voice-activity-detection","display_name":"Voice activity detection","score":0.5218150019645691},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5156453847885132},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.4894348680973053},{"id":"https://openalex.org/keywords/speech-processing","display_name":"Speech processing","score":0.4677000045776367},{"id":"https://openalex.org/keywords/waveform","display_name":"Waveform","score":0.4654463231563568},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.4611058831214905},{"id":"https://openalex.org/keywords/frame","display_name":"Frame (networking)","score":0.4565439224243164},{"id":"https://openalex.org/keywords/feature-extraction","display_name":"Feature extraction","score":0.4464240074157715},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.4261653423309326},{"id":"https://openalex.org/keywords/test-set","display_name":"Test set","score":0.4198811650276184}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7662114500999451},{"id":"https://openalex.org/C53016008","wikidata":"https://www.wikidata.org/wiki/Q620167","display_name":"Front and back ends","level":2,"score":0.6265080571174622},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5741527676582336},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.562653660774231},{"id":"https://openalex.org/C49937458","wikidata":"https://www.wikidata.org/wiki/Q2599292","display_name":"Probabilistic logic","level":2,"score":0.5477754473686218},{"id":"https://openalex.org/C204201278","wikidata":"https://www.wikidata.org/wiki/Q1332614","display_name":"Voice activity detection","level":3,"score":0.5218150019645691},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5156453847885132},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.4894348680973053},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.4677000045776367},{"id":"https://openalex.org/C197424946","wikidata":"https://www.wikidata.org/wiki/Q1165717","display_name":"Waveform","level":3,"score":0.4654463231563568},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.4611058831214905},{"id":"https://openalex.org/C126042441","wikidata":"https://www.wikidata.org/wiki/Q1324888","display_name":"Frame (networking)","level":2,"score":0.4565439224243164},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.4464240074157715},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.4261653423309326},{"id":"https://openalex.org/C169903167","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Test set","level":2,"score":0.4198811650276184},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.0},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C554190296","wikidata":"https://www.wikidata.org/wiki/Q47528","display_name":"Radar","level":2,"score":0.0},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp.2006.1660025","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp.2006.1660025","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2006 IEEE International Conference on Acoustics Speed and Signal Processing Proceedings","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.5799999833106995,"display_name":"Peace, Justice and strong institutions","id":"https://metadata.un.org/sdg/16"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":11,"referenced_works":["https://openalex.org/W131736481","https://openalex.org/W1560013842","https://openalex.org/W1838490064","https://openalex.org/W1972440476","https://openalex.org/W2071489795","https://openalex.org/W2119155265","https://openalex.org/W2125838338","https://openalex.org/W2135933868","https://openalex.org/W2151311453","https://openalex.org/W2166084932","https://openalex.org/W6605452371"],"related_works":["https://openalex.org/W642007152","https://openalex.org/W2401827384","https://openalex.org/W2355290951","https://openalex.org/W2069501481","https://openalex.org/W2052688117","https://openalex.org/W2552102772","https://openalex.org/W2103239478","https://openalex.org/W4294771049","https://openalex.org/W1523214805","https://openalex.org/W2168417340"],"abstract_inverted_index":{"In":[0],"this":[1,82],"paper":[2],"we":[3],"discuss":[4],"the":[5,10,21,86,99,102,108,137,141,144,158],"design":[6],"and":[7,29,50,119,123,157],"implementation":[8],"of":[9,27,48,59,66,77,90,105,107,135,143],"ASAT":[11,40],"front":[12,41],"end":[13,42],"processing":[14,43],"system,":[15],"whose":[16],"goal":[17],"is":[18],"to":[19,35,56,63],"convert":[20],"speech":[22,52,95],"waveform":[23],"into":[24],"a":[25,46,57,132],"range":[26,47],"measurements":[28],"parameters":[30,53],"which":[31],"are":[32,111],"then":[33],"combined":[34],"form":[36],"probabilistic":[37],"attributes.":[38,96],"The":[39],"module":[44],"utilizes":[45],"spectral":[49],"temporal":[51],"as":[54,94],"input":[55],"set":[58],"neural":[60,145],"network":[61],"classifiers":[62],"create":[64],"sets":[65],"attribute":[67],"probability":[68],"lattices,":[69],"based":[70],"on":[71],"either":[72],"single":[73],"frames":[74,78],"or":[75],"blocks":[76],"(segments).":[79],"We":[80],"test":[81],"architecture":[83],"by":[84],"using":[85],"14":[87],"Sound":[88],"Patterns":[89],"English":[91],"(SPE)":[92],"features":[93,110,115,125,151,161],"Without":[97],"balancing":[98,136],"training":[100,139],"data,":[101,140],"detection":[103,121,127,153,163],"accuracies":[104,128,154],"4":[106],"SPE":[109],"above":[112,155,165],"90%,":[113],"2":[114],"obtain":[116],"between":[117],"80%":[118,166],"90%":[120,156],"accuracy,":[122],"8":[124,160],"have":[126],"below":[129],"80%.":[130],"With":[131],"novel":[133],"method":[134],"feature":[138],"performance":[142],"networks":[146],"improved":[147],"significantly,":[148],"with":[149,162],"6":[150],"having":[152],"remaining":[159],"accuracy":[164]},"counts_by_year":[{"year":2021,"cited_by_count":1},{"year":2017,"cited_by_count":1},{"year":2014,"cited_by_count":1}],"updated_date":"2026-06-26T08:34:08.712188","created_date":"2025-10-10T00:00:00"}