{"id":"https://openalex.org/W4297678885","doi":"https://doi.org/10.1145/3549737.3549785","title":"Compressing Audio Visual Speech Recognition Models With Parameterized Hypercomplex Layers","display_name":"Compressing Audio Visual Speech Recognition Models With Parameterized Hypercomplex Layers","publication_year":2022,"publication_date":"2022-09-07","ids":{"openalex":"https://openalex.org/W4297678885","doi":"https://doi.org/10.1145/3549737.3549785"},"language":"en","primary_location":{"id":"doi:10.1145/3549737.3549785","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3549737.3549785","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 12th Hellenic Conference on Artificial Intelligence","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/A5065704831","display_name":"Iason Ioannis Panagos","orcid":null},"institutions":[{"id":"https://openalex.org/I194019607","display_name":"University of Ioannina","ror":"https://ror.org/01qg3j183","country_code":"GR","type":"education","lineage":["https://openalex.org/I194019607"]}],"countries":["GR"],"is_corresponding":true,"raw_author_name":"Iason Ioannis Panagos","raw_affiliation_strings":["Department of Computer Science & Engineering, University of Ioannina, Greece"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science & Engineering, University of Ioannina, Greece","institution_ids":["https://openalex.org/I194019607"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5039126972","display_name":"Giorgos Sfikas","orcid":"https://orcid.org/0000-0002-7305-2886"},"institutions":[{"id":"https://openalex.org/I194019607","display_name":"University of Ioannina","ror":"https://ror.org/01qg3j183","country_code":"GR","type":"education","lineage":["https://openalex.org/I194019607"]}],"countries":["GR"],"is_corresponding":false,"raw_author_name":"Giorgos Sfikas","raw_affiliation_strings":["Computational Intelligence Laboratory, Institute of Informatics and Telecommunications, National Center for Scientific Research, Greece and Department of Computer Science & Engineering, University of Ioannina, Greece"],"affiliations":[{"raw_affiliation_string":"Computational Intelligence Laboratory, Institute of Informatics and Telecommunications, National Center for Scientific Research, Greece and Department of Computer Science & Engineering, University of Ioannina, Greece","institution_ids":["https://openalex.org/I194019607"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5058318737","display_name":"Christophoros Nikou","orcid":"https://orcid.org/0000-0003-1388-6915"},"institutions":[{"id":"https://openalex.org/I194019607","display_name":"University of Ioannina","ror":"https://ror.org/01qg3j183","country_code":"GR","type":"education","lineage":["https://openalex.org/I194019607"]}],"countries":["GR"],"is_corresponding":false,"raw_author_name":"Christophoros Nikou","raw_affiliation_strings":["Department of Computer Science & Engineering, University of Ioannina, Greece"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science & Engineering, University of Ioannina, Greece","institution_ids":["https://openalex.org/I194019607"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5065704831"],"corresponding_institution_ids":["https://openalex.org/I194019607"],"apc_list":null,"apc_paid":null,"fwci":0.7437,"has_fulltext":false,"cited_by_count":5,"citation_normalized_percentile":{"value":0.68776336,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"7"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":1.0,"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"}},"topics":[{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":1.0,"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":0.9998000264167786,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.9991000294685364,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8057416677474976},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.7573323249816895},{"id":"https://openalex.org/keywords/parameterized-complexity","display_name":"Parameterized complexity","score":0.6112518906593323},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.610435962677002},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5872396230697632},{"id":"https://openalex.org/keywords/variety","display_name":"Variety (cybernetics)","score":0.5690492391586304},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.5162803530693054},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4534362554550171},{"id":"https://openalex.org/keywords/recurrent-neural-network","display_name":"Recurrent neural network","score":0.4458417296409607},{"id":"https://openalex.org/keywords/resource","display_name":"Resource (disambiguation)","score":0.43559902906417847},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3774411678314209},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.37333551049232483}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8057416677474976},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.7573323249816895},{"id":"https://openalex.org/C165464430","wikidata":"https://www.wikidata.org/wiki/Q1570441","display_name":"Parameterized complexity","level":2,"score":0.6112518906593323},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.610435962677002},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5872396230697632},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.5690492391586304},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.5162803530693054},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4534362554550171},{"id":"https://openalex.org/C147168706","wikidata":"https://www.wikidata.org/wiki/Q1457734","display_name":"Recurrent neural network","level":3,"score":0.4458417296409607},{"id":"https://openalex.org/C206345919","wikidata":"https://www.wikidata.org/wiki/Q20380951","display_name":"Resource (disambiguation)","level":2,"score":0.43559902906417847},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3774411678314209},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.37333551049232483},{"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/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.0},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3549737.3549785","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3549737.3549785","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 12th Hellenic Conference on Artificial Intelligence","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":19,"referenced_works":["https://openalex.org/W2015143272","https://openalex.org/W2096391593","https://openalex.org/W2404704342","https://openalex.org/W2550143307","https://openalex.org/W2787285847","https://openalex.org/W2894977338","https://openalex.org/W2952383053","https://openalex.org/W2962912639","https://openalex.org/W2963030892","https://openalex.org/W2963192365","https://openalex.org/W2963230471","https://openalex.org/W2963654155","https://openalex.org/W2972756321","https://openalex.org/W3016011581","https://openalex.org/W3081924603","https://openalex.org/W3128564814","https://openalex.org/W3162707322","https://openalex.org/W3167917117","https://openalex.org/W4210856350"],"related_works":["https://openalex.org/W2972825155","https://openalex.org/W2891300059","https://openalex.org/W4226493464","https://openalex.org/W4312417841","https://openalex.org/W3133861977","https://openalex.org/W3008584592","https://openalex.org/W2951211570","https://openalex.org/W3103566983","https://openalex.org/W3167935049","https://openalex.org/W3029198973"],"abstract_inverted_index":{"Audio":[0],"visual":[1],"speech":[2,75,140],"recognition":[3,141],"has":[4],"seen":[5],"remarkable":[6],"progress":[7,14],"in":[8,24,71,78,105,139],"the":[9,19,37,72,82,113,118,150,162],"last":[10],"few":[11],"years.":[12],"This":[13],"is":[15,84,112],"a":[16,50],"result,":[17],"on":[18,36],"one":[20],"hand,":[21,39],"of":[22,53,74,91,120,136,152],"advances":[23],"deep":[25,62],"learning-based":[26],"architectures,":[27],"such":[28],"as":[29],"convolutional":[30],"and":[31,97,100,126],"recurrent":[32],"neural":[33],"networks,":[34],"and,":[35],"other":[38],"due":[40,116],"to":[41,60,117,133],"large-scale":[42],"public":[43],"datasets":[44],"have":[45,57],"been":[46],"introduced":[47],"that":[48,64,68,93,148,158],"provide":[49],"great":[51],"variety":[52],"speakers.":[54],"Both":[55],"factors":[56],"led":[58],"authors":[59],"develop":[61],"architectures":[63,88],"achieve":[65],"impressive":[66],"results":[67],"surpass":[69],"humans":[70],"areas":[73],"recognition,":[76],"especially":[77],"cases":[79],"where":[80],"only":[81],"video":[83],"present.":[85],"Nevertheless,":[86],"these":[87,137],"involve":[89],"millions":[90],"parameters":[92],"increase":[94],"their":[95,103],"storage":[96],"memory":[98],"demands":[99],"also":[101],"limit":[102],"deployment":[104],"resource":[106],"constrained":[107],"scenarios.":[108],"An":[109],"additional":[110],"issue":[111],"energy":[114],"expenditure":[115],"amount":[119],"calculations":[121],"required":[122,153],"for":[123],"training,":[124],"fine-tuning":[125],"testing.":[127],"In":[128],"this":[129],"work,":[130],"we":[131],"attempt":[132],"mitigate":[134],"some":[135],"shortcomings":[138],"models":[142,157],"by":[143],"incorporating":[144],"parameterized":[145],"hypercomplex":[146],"layers":[147],"reduce":[149],"number":[151],"resources.":[154],"We":[155],"present":[156],"are":[159],"competitive":[160],"with":[161,166],"state-of-the-art":[163],"while":[164],"operating":[165],"fewer":[167],"parameters.":[168]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":3},{"year":2023,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}