{"id":"https://openalex.org/W3163570686","doi":"https://doi.org/10.1109/icassp39728.2021.9413630","title":"Neural Architecture Search for LF-MMI Trained Time Delay Neural Networks","display_name":"Neural Architecture Search for LF-MMI Trained Time Delay Neural Networks","publication_year":2021,"publication_date":"2021-05-13","ids":{"openalex":"https://openalex.org/W3163570686","doi":"https://doi.org/10.1109/icassp39728.2021.9413630","mag":"3163570686"},"language":"en","primary_location":{"id":"doi:10.1109/icassp39728.2021.9413630","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9413630","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5045355404","display_name":"Shoukang Hu","orcid":"https://orcid.org/0000-0002-3345-6923"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Shoukang Hu","raw_affiliation_strings":["The Chinese University of Hong Kong,Hong Kong SAR,China","The Chinese University of Hong Kong, Hong Kong SAR, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong,Hong Kong SAR,China","institution_ids":["https://openalex.org/I177725633"]},{"raw_affiliation_string":"The Chinese University of Hong Kong, Hong Kong SAR, China","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5113476082","display_name":"Xurong Xie","orcid":"https://orcid.org/0009-0007-6729-2879"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Xurong Xie","raw_affiliation_strings":["The Chinese University of Hong Kong,Hong Kong SAR,China","The Chinese University of Hong Kong, Hong Kong SAR, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong,Hong Kong SAR,China","institution_ids":["https://openalex.org/I177725633"]},{"raw_affiliation_string":"The Chinese University of Hong Kong, Hong Kong SAR, China","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066070154","display_name":"Shansong Liu","orcid":"https://orcid.org/0000-0001-6202-5615"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Shansong Liu","raw_affiliation_strings":["The Chinese University of Hong Kong,Hong Kong SAR,China","The Chinese University of Hong Kong, Hong Kong SAR, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong,Hong Kong SAR,China","institution_ids":["https://openalex.org/I177725633"]},{"raw_affiliation_string":"The Chinese University of Hong Kong, Hong Kong SAR, China","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5065896539","display_name":"Mingyu Cui","orcid":"https://orcid.org/0000-0001-9781-9407"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Mingyu Cui","raw_affiliation_strings":["The Chinese University of Hong Kong,Hong Kong SAR,China","The Chinese University of Hong Kong, Hong Kong SAR, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong,Hong Kong SAR,China","institution_ids":["https://openalex.org/I177725633"]},{"raw_affiliation_string":"The Chinese University of Hong Kong, Hong Kong SAR, China","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003895235","display_name":"Mengzhe Geng","orcid":"https://orcid.org/0000-0002-7886-439X"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Mengzhe Geng","raw_affiliation_strings":["The Chinese University of Hong Kong,Hong Kong SAR,China","The Chinese University of Hong Kong, Hong Kong SAR, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong,Hong Kong SAR,China","institution_ids":["https://openalex.org/I177725633"]},{"raw_affiliation_string":"The Chinese University of Hong Kong, Hong Kong SAR, China","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5037109470","display_name":"Xunying Liu","orcid":"https://orcid.org/0000-0001-6725-1160"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Xunying Liu","raw_affiliation_strings":["The Chinese University of Hong Kong,Hong Kong SAR,China","The Chinese University of Hong Kong, Hong Kong SAR, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong,Hong Kong SAR,China","institution_ids":["https://openalex.org/I177725633"]},{"raw_affiliation_string":"The Chinese University of Hong Kong, Hong Kong SAR, China","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5019458385","display_name":"Helen Meng","orcid":"https://orcid.org/0000-0002-4427-3532"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Helen Meng","raw_affiliation_strings":["The Chinese University of Hong Kong,Hong Kong SAR,China","The Chinese University of Hong Kong, Hong Kong SAR, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong,Hong Kong SAR,China","institution_ids":["https://openalex.org/I177725633"]},{"raw_affiliation_string":"The Chinese University of Hong Kong, Hong Kong SAR, China","institution_ids":["https://openalex.org/I177725633"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I177725633"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":6,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"6758","last_page":"6762"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9997000098228455,"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":0.9997000098228455,"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":0.9987000226974487,"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":0.9965999722480774,"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":0.8201022148132324},{"id":"https://openalex.org/keywords/softmax-function","display_name":"Softmax function","score":0.76934415102005},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5787920951843262},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5426235198974609},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5385144352912903},{"id":"https://openalex.org/keywords/word-error-rate","display_name":"Word error rate","score":0.5145427584648132},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.5017352104187012},{"id":"https://openalex.org/keywords/time-delay-neural-network","display_name":"Time delay neural network","score":0.47954320907592773},{"id":"https://openalex.org/keywords/network-architecture","display_name":"Network architecture","score":0.42328551411628723},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.37187057733535767},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.327402800321579}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8201022148132324},{"id":"https://openalex.org/C188441871","wikidata":"https://www.wikidata.org/wiki/Q7554146","display_name":"Softmax function","level":3,"score":0.76934415102005},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5787920951843262},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5426235198974609},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5385144352912903},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.5145427584648132},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5017352104187012},{"id":"https://openalex.org/C175202392","wikidata":"https://www.wikidata.org/wiki/Q2434543","display_name":"Time delay neural network","level":3,"score":0.47954320907592773},{"id":"https://openalex.org/C193415008","wikidata":"https://www.wikidata.org/wiki/Q639681","display_name":"Network architecture","level":2,"score":0.42328551411628723},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.37187057733535767},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.327402800321579},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C151730666","wikidata":"https://www.wikidata.org/wiki/Q7205","display_name":"Paleontology","level":1,"score":0.0},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp39728.2021.9413630","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9413630","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":61,"referenced_works":["https://openalex.org/W180052447","https://openalex.org/W1491129255","https://openalex.org/W1524333225","https://openalex.org/W1527646298","https://openalex.org/W1987238397","https://openalex.org/W2002342963","https://openalex.org/W2111935653","https://openalex.org/W2114016253","https://openalex.org/W2131342762","https://openalex.org/W2146871184","https://openalex.org/W2167277568","https://openalex.org/W2169163929","https://openalex.org/W2402146185","https://openalex.org/W2514741789","https://openalex.org/W2548228487","https://openalex.org/W2553303224","https://openalex.org/W2788853733","https://openalex.org/W2810075754","https://openalex.org/W2885311373","https://openalex.org/W2888867175","https://openalex.org/W2905672847","https://openalex.org/W2925303509","https://openalex.org/W2932077855","https://openalex.org/W2951104886","https://openalex.org/W2952165242","https://openalex.org/W2962746461","https://openalex.org/W2962750597","https://openalex.org/W2962847160","https://openalex.org/W2963032572","https://openalex.org/W2963037463","https://openalex.org/W2963136578","https://openalex.org/W2964212578","https://openalex.org/W2964259004","https://openalex.org/W2972482531","https://openalex.org/W2972581070","https://openalex.org/W2997326985","https://openalex.org/W3024022091","https://openalex.org/W3034309359","https://openalex.org/W3082696729","https://openalex.org/W3095123370","https://openalex.org/W3095175338","https://openalex.org/W3096329809","https://openalex.org/W3096533519","https://openalex.org/W3097909406","https://openalex.org/W3157924851","https://openalex.org/W4300687381","https://openalex.org/W6631362777","https://openalex.org/W6712930963","https://openalex.org/W6729956949","https://openalex.org/W6730091202","https://openalex.org/W6748057086","https://openalex.org/W6748587240","https://openalex.org/W6752515464","https://openalex.org/W6753278433","https://openalex.org/W6756887525","https://openalex.org/W6757204547","https://openalex.org/W6760796423","https://openalex.org/W6761158446","https://openalex.org/W6766029800","https://openalex.org/W6772281271","https://openalex.org/W6782969793"],"related_works":["https://openalex.org/W2309273277","https://openalex.org/W2061937230","https://openalex.org/W1769849273","https://openalex.org/W1574295218","https://openalex.org/W113247760","https://openalex.org/W1967477266","https://openalex.org/W2547793174","https://openalex.org/W2070212102","https://openalex.org/W2132658536","https://openalex.org/W2544241817"],"abstract_inverted_index":{"Deep":[0],"neural":[1,25,44],"networks":[2,45],"(DNNs)":[3],"based":[4],"automatic":[5],"speech":[6,189],"recognition":[7,190],"(ASR)":[8],"systems":[9,138,145],"are":[10,30],"often":[11],"designed":[12],"using":[13,146,192],"expert":[14],"knowledge":[15],"and":[16,50,55,83,92,99,111,158,170],"empirical":[17],"evaluation.":[18],"In":[19],"this":[20],"paper,":[21],"a":[22,186],"range":[23],"of":[24,37,39,59,96,175],"architecture":[26,74,97,152],"search":[27,121,153],"(NAS)":[28],"techniques":[29],"used":[31],"to":[32,105,124,168],"automatically":[33],"learn":[34],"two":[35],"types":[36],"hyper-parameters":[38],"state-of-the-art":[40],"factored":[41],"time":[42],"delay":[43],"(TDNNs):":[46],"i)":[47],"the":[48,57,60,70,87,94,107,131,136,141,193],"left":[49],"right":[51],"splicing":[52],"context":[53],"offsets;":[54],"ii)":[56],"dimensionality":[58],"bottleneck":[61],"linear":[62],"projection":[63],"at":[64],"each":[65],"hidden":[66],"layer.":[67],"These":[68],"include":[69],"DARTS":[71,85,101],"method":[72],"integrating":[73],"selection":[75],"with":[76],"lattice-free":[77],"MMI":[78],"(LF-MMI)":[79],"TDNN":[80,126,144],"training;":[81],"Gumbel-Softmax":[82],"pipelined":[84],"reducing":[86],"confusion":[88],"over":[89,122],"candidate":[90,117],"architectures":[91,118],"improving":[93],"generalization":[95],"selection;":[98],"Penalized":[100],"incorporating":[102],"resource":[103],"constraints":[104],"adjust":[106],"trade-off":[108],"between":[109],"performance":[110,180],"system":[112],"complexity.":[113],"Parameter":[114],"sharing":[115],"among":[116],"allows":[119],"efficient":[120],"up":[123,167],"728different":[125],"systems.":[127],"Experiments":[128],"conducted":[129],"on":[130,185],"300-hour":[132],"Switchboard":[133],"corpus":[134],"suggest":[135],"auto-configured":[137],"consistently":[139],"outperform":[140],"baseline":[142],"LF-MMI":[143],"manual":[147],"network":[148],"design":[149],"or":[150],"random":[151],"after":[154],"LHUC":[155],"speaker":[156],"adaptation":[157],"RNNLM":[159],"rescoring.":[160],"Absolute":[161],"word":[162],"error":[163],"rate":[164],"(WER)":[165],"reductions":[166],"1.0%":[169],"relative":[171],"model":[172],"size":[173],"reduction":[174],"28%":[176],"were":[177,182],"obtained.":[178],"Consistent":[179],"improvements":[181],"also":[183],"obtained":[184],"UASpeech":[187],"disordered":[188],"task":[191],"proposed":[194],"NAS":[195],"approaches.":[196]},"counts_by_year":[{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":1},{"year":2022,"cited_by_count":2},{"year":2021,"cited_by_count":2}],"updated_date":"2026-07-29T09:40:50.615796","created_date":"2025-10-10T00:00:00"}