{"id":"https://openalex.org/W4408347312","doi":"https://doi.org/10.1109/icassp49660.2025.10889844","title":"Linear Time Complexity Conformers with SummaryMixing for Streaming Speech Recognition","display_name":"Linear Time Complexity Conformers with SummaryMixing for Streaming Speech Recognition","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408347312","doi":"https://doi.org/10.1109/icassp49660.2025.10889844"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10889844","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10889844","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2025 - 2025 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/A5089505434","display_name":"Titouan Parcollet","orcid":"https://orcid.org/0000-0003-0672-1346"},"institutions":[{"id":"https://openalex.org/I4210117523","display_name":"Samsung (United Kingdom)","ror":"https://ror.org/01w6gjq94","country_code":"GB","type":"company","lineage":["https://openalex.org/I2250650973","https://openalex.org/I4210117523"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Titouan Parcollet","raw_affiliation_strings":["Samsung AI Center – Cambridge,Cambridge,United Kingdom"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung AI Center – Cambridge,Cambridge,United Kingdom","institution_ids":["https://openalex.org/I4210117523"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5045282611","display_name":"Rogier van Dalen","orcid":null},"institutions":[{"id":"https://openalex.org/I4210117523","display_name":"Samsung (United Kingdom)","ror":"https://ror.org/01w6gjq94","country_code":"GB","type":"company","lineage":["https://openalex.org/I2250650973","https://openalex.org/I4210117523"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Rogier van Dalen","raw_affiliation_strings":["Samsung AI Center – Cambridge,Cambridge,United Kingdom"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung AI Center – Cambridge,Cambridge,United Kingdom","institution_ids":["https://openalex.org/I4210117523"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003030422","display_name":"Shucong Zhang","orcid":null},"institutions":[{"id":"https://openalex.org/I4210117523","display_name":"Samsung (United Kingdom)","ror":"https://ror.org/01w6gjq94","country_code":"GB","type":"company","lineage":["https://openalex.org/I2250650973","https://openalex.org/I4210117523"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Shucong Zhang","raw_affiliation_strings":["Samsung AI Center – Cambridge,Cambridge,United Kingdom"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung AI Center – Cambridge,Cambridge,United Kingdom","institution_ids":["https://openalex.org/I4210117523"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5087701164","display_name":"Sourav Bhattacharya","orcid":"https://orcid.org/0000-0001-9670-5264"},"institutions":[{"id":"https://openalex.org/I4210117523","display_name":"Samsung (United Kingdom)","ror":"https://ror.org/01w6gjq94","country_code":"GB","type":"company","lineage":["https://openalex.org/I2250650973","https://openalex.org/I4210117523"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Sourav Bhattacharya","raw_affiliation_strings":["Samsung AI Center – Cambridge,Cambridge,United Kingdom"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung AI Center – Cambridge,Cambridge,United Kingdom","institution_ids":["https://openalex.org/I4210117523"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210117523"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"5"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10901","display_name":"Advanced Data Compression Techniques","score":0.9909999966621399,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T10901","display_name":"Advanced Data Compression Techniques","score":0.9909999966621399,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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":0.9896000027656555,"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":0.9811000227928162,"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":0.6970273852348328},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5380145907402039},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3534868359565735}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6970273852348328},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5380145907402039},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3534868359565735}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49660.2025.10889844","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10889844","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2025 - 2025 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":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":27,"referenced_works":["https://openalex.org/W108866686","https://openalex.org/W1494198834","https://openalex.org/W2245493112","https://openalex.org/W2521054263","https://openalex.org/W2912581782","https://openalex.org/W2962780374","https://openalex.org/W3085139254","https://openalex.org/W3097777922","https://openalex.org/W3142589538","https://openalex.org/W3160766462","https://openalex.org/W3162665866","https://openalex.org/W3167533889","https://openalex.org/W3197478142","https://openalex.org/W3197813307","https://openalex.org/W4319862255","https://openalex.org/W4372346432","https://openalex.org/W4385245566","https://openalex.org/W4402112115","https://openalex.org/W4402112522","https://openalex.org/W6761628794","https://openalex.org/W6779163297","https://openalex.org/W6788158735","https://openalex.org/W6800388583","https://openalex.org/W6838276489","https://openalex.org/W6839026989","https://openalex.org/W6847363464","https://openalex.org/W6857062747"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"Automatic":[0],"speech":[1,22,54,80,108],"recognition":[2,55],"(ASR)":[3],"with":[4,8],"an":[5,98],"encoder":[6,109],"equipped":[7],"self-attention,":[9],"whether":[10],"streaming":[11,79,96],"or":[12,62],"non-streaming,":[13],"takes":[14],"quadratic":[15],"time":[16],"in":[17,40,93,112],"the":[18,21,31,35,38,58,64,70],"length":[19],"of":[20,37,66,73],"utterance.":[23],"This":[24],"slows":[25],"down":[26],"training":[27,122],"and":[28,33,97,119,123],"decoding,":[29],"increase":[30],"cost,":[32],"limits":[34],"deployment":[36],"ASR":[39],"constrained":[41],"devices.":[42],"SummaryMixing":[43,74,86],"is":[44,75],"a":[45,88,95],"promising":[46],"linear-time":[47,106],"complexity":[48,107],"alternative":[49],"to":[50,78,87],"self-attention":[51,67,111],"for":[52,57],"non-streaming":[53],"that,":[56],"first":[59],"time,":[60],"preserves":[61],"outperforms":[63,110],"accuracy":[65],"models.":[68],"Unfortunately,":[69],"original":[71],"definition":[72],"not":[76],"suited":[77],"recognition.":[81],"Hence,":[82],"this":[83,104],"work":[84],"extends":[85],"Conformer":[89],"Transducer":[90],"that":[91,103],"works":[92],"both":[94,113],"offline":[99],"mode.":[100],"It":[101],"shows":[102],"new":[105],"scenarios":[114],"while":[115],"requiring":[116],"less":[117],"compute":[118],"memory":[120],"during":[121],"decoding.":[124]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}