{"id":"https://openalex.org/W3169722760","doi":"https://doi.org/10.21437/interspeech.2021-655","title":"Multi-Channel Transformer Transducer for Speech Recognition","display_name":"Multi-Channel Transformer Transducer for Speech Recognition","publication_year":2021,"publication_date":"2021-08-27","ids":{"openalex":"https://openalex.org/W3169722760","doi":"https://doi.org/10.21437/interspeech.2021-655","mag":"3169722760"},"language":"en","primary_location":{"id":"doi:10.21437/interspeech.2021-655","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2021-655","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2021","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/A5102963746","display_name":"Feng-Ju Chang","orcid":"https://orcid.org/0000-0003-2405-3118"},"institutions":[{"id":"https://openalex.org/I1311688040","display_name":"Amazon (United States)","ror":"https://ror.org/04mv4n011","country_code":"US","type":"company","lineage":["https://openalex.org/I1311688040"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Feng-Ju Chang","raw_affiliation_strings":["Alexa Machine Learning, Amazon, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alexa Machine Learning, Amazon, USA","institution_ids":["https://openalex.org/I1311688040"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5023415990","display_name":"Martin Radfar","orcid":null},"institutions":[{"id":"https://openalex.org/I1311688040","display_name":"Amazon (United States)","ror":"https://ror.org/04mv4n011","country_code":"US","type":"company","lineage":["https://openalex.org/I1311688040"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Martin Radfar","raw_affiliation_strings":["Alexa Machine Learning, Amazon, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alexa Machine Learning, Amazon, USA","institution_ids":["https://openalex.org/I1311688040"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029573067","display_name":"Athanasios Mouchtaris","orcid":"https://orcid.org/0000-0001-7583-0189"},"institutions":[{"id":"https://openalex.org/I1311688040","display_name":"Amazon (United States)","ror":"https://ror.org/04mv4n011","country_code":"US","type":"company","lineage":["https://openalex.org/I1311688040"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Athanasios Mouchtaris","raw_affiliation_strings":["Alexa Machine Learning, Amazon, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alexa Machine Learning, Amazon, USA","institution_ids":["https://openalex.org/I1311688040"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5025736898","display_name":"Maurizio Omologo","orcid":"https://orcid.org/0000-0003-0879-0548"},"institutions":[{"id":"https://openalex.org/I1311688040","display_name":"Amazon (United States)","ror":"https://ror.org/04mv4n011","country_code":"US","type":"company","lineage":["https://openalex.org/I1311688040"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Maurizio Omologo","raw_affiliation_strings":["Alexa Machine Learning, Amazon, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Alexa Machine Learning, Amazon, USA","institution_ids":["https://openalex.org/I1311688040"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I1311688040"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":23,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"296","last_page":"300"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9976999759674072,"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":0.9976999759674072,"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":0.9976999759674072,"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/T10662","display_name":"Ultrasonics and Acoustic Wave Propagation","score":0.9812999963760376,"subfield":{"id":"https://openalex.org/subfields/2211","display_name":"Mechanics of Materials"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/transducer","display_name":"Transducer","score":0.7441050410270691},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6194764375686646},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.5856631398200989},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5321005582809448},{"id":"https://openalex.org/keywords/acoustics","display_name":"Acoustics","score":0.395113468170166},{"id":"https://openalex.org/keywords/electronic-engineering","display_name":"Electronic engineering","score":0.33568233251571655},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.30385202169418335},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.23808729648590088},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.1551360785961151},{"id":"https://openalex.org/keywords/physics","display_name":"Physics","score":0.09309864044189453}],"concepts":[{"id":"https://openalex.org/C56318395","wikidata":"https://www.wikidata.org/wiki/Q215928","display_name":"Transducer","level":2,"score":0.7441050410270691},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6194764375686646},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5856631398200989},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5321005582809448},{"id":"https://openalex.org/C24890656","wikidata":"https://www.wikidata.org/wiki/Q82811","display_name":"Acoustics","level":1,"score":0.395113468170166},{"id":"https://openalex.org/C24326235","wikidata":"https://www.wikidata.org/wiki/Q126095","display_name":"Electronic engineering","level":1,"score":0.33568233251571655},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.30385202169418335},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.23808729648590088},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.1551360785961151},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.09309864044189453}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.21437/interspeech.2021-655","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2021-655","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2021","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":41,"referenced_works":["https://openalex.org/W81726370","https://openalex.org/W1522301498","https://openalex.org/W1828163288","https://openalex.org/W1982728343","https://openalex.org/W1985368711","https://openalex.org/W2109799268","https://openalex.org/W2129227482","https://openalex.org/W2133564696","https://openalex.org/W2289394825","https://openalex.org/W2398042854","https://openalex.org/W2398972335","https://openalex.org/W2490695385","https://openalex.org/W2506203739","https://openalex.org/W2517616541","https://openalex.org/W2636483419","https://openalex.org/W2892009249","https://openalex.org/W2892163332","https://openalex.org/W2900893004","https://openalex.org/W2918296821","https://openalex.org/W2921214093","https://openalex.org/W2954695182","https://openalex.org/W2962760690","https://openalex.org/W2962784628","https://openalex.org/W2962970794","https://openalex.org/W2963399332","https://openalex.org/W2976556660","https://openalex.org/W2981857663","https://openalex.org/W3008762051","https://openalex.org/W3015194534","https://openalex.org/W3015834770","https://openalex.org/W3016010032","https://openalex.org/W3086154751","https://openalex.org/W3096686110","https://openalex.org/W3097777922","https://openalex.org/W3097973766","https://openalex.org/W3124547804","https://openalex.org/W3160207687","https://openalex.org/W4206633762","https://openalex.org/W4288088457","https://openalex.org/W4294619417","https://openalex.org/W4385245566"],"related_works":["https://openalex.org/W2072454424","https://openalex.org/W2383068986","https://openalex.org/W2090275562","https://openalex.org/W3152005513","https://openalex.org/W2020960496","https://openalex.org/W1266100159","https://openalex.org/W2034887142","https://openalex.org/W2088920587","https://openalex.org/W2139162221","https://openalex.org/W2092329823"],"abstract_inverted_index":{"Multi-channel":[0],"inputs":[1,27],"offer":[2],"several":[3],"advantages":[4],"over":[5],"singlechannel,":[6],"to":[7,24,101,114],"improve":[8,132],"the":[9,110,133,140],"robustness":[10],"of":[11,124,136],"on-device":[12,50,85],"speech":[13,58,86],"recognition":[14,59],"systems.Recent":[15],"work":[16],"on":[17],"multi-channel":[18,68,96,111],"transformer,":[19],"has":[20],"proposed":[21],"a":[22,39,56,88],"way":[23],"incorporate":[25],"such":[26],"into":[28],"end-to-end":[29,67],"ASR":[30],"for":[31,81],"improved":[32],"accuracy.However,":[33],"this":[34,52],"approach":[35],"is":[36,79,118],"characterized":[37],"by":[38,138],"high":[40],"computational":[41,134],"complexity,":[42],"which":[43,65],"prevents":[44],"it":[45,78],"from":[46],"being":[47],"deployed":[48],"in":[49,84,122,145],"systems.In":[51],"paper,":[53],"we":[54,130],"present":[55],"novel":[57],"model,":[60],"Multi-Channel":[61],"Transformer":[62],"Transducer":[63],"(MCTT),":[64],"features":[66],"training,":[69],"low":[70,74],"computation":[71],"cost,":[72],"and":[73,117,142],"latency":[75],"so":[76],"that":[77,129],"suitable":[80],"streaming":[82],"decoding":[83],"recognition.In":[87],"far-field":[89],"in-house":[90],"dataset,":[91],"our":[92],"MCTT":[93,108,137],"outperforms":[94,109],"stagewise":[95],"models":[97],"with":[98],"transformer-transducer":[99],"up":[100,113],"6.01%":[102],"relative":[103],"WER":[104],"improvement":[105],"(WERR).In":[106],"addition,":[107],"transformer":[112],"11.62%":[115],"WERR,":[116],"15.8":[119],"times":[120],"faster":[121],"terms":[123],"inference":[125],"speed.We":[126],"further":[127],"show":[128],"can":[131],"cost":[135],"constraining":[139],"future":[141],"previous":[143],"context":[144],"attention":[146],"computations.":[147]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":6},{"year":2023,"cited_by_count":5},{"year":2022,"cited_by_count":6},{"year":2021,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}