{"id":"https://openalex.org/W4408354546","doi":"https://doi.org/10.1109/icassp49660.2025.10889296","title":"SimulTron: On-Device Simultaneous Speech to Speech Translation","display_name":"SimulTron: On-Device Simultaneous Speech to Speech Translation","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408354546","doi":"https://doi.org/10.1109/icassp49660.2025.10889296"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10889296","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10889296","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":"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/A5099043227","display_name":"Alex Agranovich","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Alex Agranovich","raw_affiliation_strings":["Google Research"],"affiliations":[{"raw_affiliation_string":"Google Research","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5083811924","display_name":"Eliya Nachmani","orcid":"https://orcid.org/0000-0003-4220-5672"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Eliya Nachmani","raw_affiliation_strings":["Google Research"],"affiliations":[{"raw_affiliation_string":"Google Research","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5070398306","display_name":"Oleg Rybakov\u200e","orcid":"https://orcid.org/0000-0003-4805-3083"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Oleg Rybakov","raw_affiliation_strings":["Google Research"],"affiliations":[{"raw_affiliation_string":"Google Research","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5107006573","display_name":"Yifan Ding","orcid":"https://orcid.org/0009-0001-4352-5427"},"institutions":[{"id":"https://openalex.org/I4210090411","display_name":"DeepMind (United Kingdom)","ror":"https://ror.org/00971b260","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210090411","https://openalex.org/I4210128969"]},{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["GB","US"],"is_corresponding":false,"raw_author_name":"Yifan Ding","raw_affiliation_strings":["Google DeepMind"],"affiliations":[{"raw_affiliation_string":"Google DeepMind","institution_ids":["https://openalex.org/I1291425158","https://openalex.org/I4210090411"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101778490","display_name":"Jia Ye","orcid":"https://orcid.org/0000-0002-7242-6065"},"institutions":[{"id":"https://openalex.org/I4210090411","display_name":"DeepMind (United Kingdom)","ror":"https://ror.org/00971b260","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210090411","https://openalex.org/I4210128969"]},{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["GB","US"],"is_corresponding":false,"raw_author_name":"Ye Jia","raw_affiliation_strings":["Google DeepMind"],"affiliations":[{"raw_affiliation_string":"Google DeepMind","institution_ids":["https://openalex.org/I1291425158","https://openalex.org/I4210090411"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5006967611","display_name":"Nadav Bar","orcid":"https://orcid.org/0000-0002-4475-8003"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Nadav Bar","raw_affiliation_strings":["Google Research"],"affiliations":[{"raw_affiliation_string":"Google Research","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003420204","display_name":"Heiga Zen","orcid":"https://orcid.org/0000-0002-8959-5471"},"institutions":[{"id":"https://openalex.org/I4210090411","display_name":"DeepMind (United Kingdom)","ror":"https://ror.org/00971b260","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210090411","https://openalex.org/I4210128969"]},{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["GB","US"],"is_corresponding":false,"raw_author_name":"Heiga Zen","raw_affiliation_strings":["Google DeepMind"],"affiliations":[{"raw_affiliation_string":"Google DeepMind","institution_ids":["https://openalex.org/I1291425158","https://openalex.org/I4210090411"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5044908343","display_name":"Michelle Tadmor Ramanovich","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Michelle Tadmor Ramanovich","raw_affiliation_strings":["Google Research"],"affiliations":[{"raw_affiliation_string":"Google Research","institution_ids":["https://openalex.org/I1291425158"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":8,"corresponding_author_ids":["https://openalex.org/A5099043227"],"corresponding_institution_ids":["https://openalex.org/I1291425158"],"apc_list":null,"apc_paid":null,"fwci":5.6828,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.9487431,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":95,"max":96},"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.8787999749183655,"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":0.8787999749183655,"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/T12031","display_name":"Speech and dialogue systems","score":0.8579999804496765,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.7774999737739563,"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/speech-translation","display_name":"Speech translation","score":0.8014233112335205},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7216968536376953},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.6698654890060425},{"id":"https://openalex.org/keywords/translation","display_name":"Translation (biology)","score":0.49300143122673035},{"id":"https://openalex.org/keywords/speech-synthesis","display_name":"Speech synthesis","score":0.4571109414100647},{"id":"https://openalex.org/keywords/speech-technology","display_name":"Speech technology","score":0.45241376757621765},{"id":"https://openalex.org/keywords/machine-translation","display_name":"Machine translation","score":0.31051549315452576},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.30475085973739624}],"concepts":[{"id":"https://openalex.org/C2780366754","wikidata":"https://www.wikidata.org/wiki/Q7494857","display_name":"Speech translation","level":3,"score":0.8014233112335205},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7216968536376953},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.6698654890060425},{"id":"https://openalex.org/C149364088","wikidata":"https://www.wikidata.org/wiki/Q185917","display_name":"Translation (biology)","level":4,"score":0.49300143122673035},{"id":"https://openalex.org/C14999030","wikidata":"https://www.wikidata.org/wiki/Q16346","display_name":"Speech synthesis","level":2,"score":0.4571109414100647},{"id":"https://openalex.org/C504749915","wikidata":"https://www.wikidata.org/wiki/Q9010971","display_name":"Speech technology","level":3,"score":0.45241376757621765},{"id":"https://openalex.org/C203005215","wikidata":"https://www.wikidata.org/wiki/Q79798","display_name":"Machine translation","level":2,"score":0.31051549315452576},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.30475085973739624},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C105580179","wikidata":"https://www.wikidata.org/wiki/Q188928","display_name":"Messenger RNA","level":3,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49660.2025.10889296","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10889296","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":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":34,"referenced_works":["https://openalex.org/W2097203679","https://openalex.org/W2136545725","https://openalex.org/W2951456627","https://openalex.org/W2964161387","https://openalex.org/W2972495969","https://openalex.org/W3007068036","https://openalex.org/W3092085609","https://openalex.org/W3175871055","https://openalex.org/W3180374548","https://openalex.org/W3209059054","https://openalex.org/W3215615641","https://openalex.org/W4226033575","https://openalex.org/W4287854499","https://openalex.org/W4297841754","https://openalex.org/W4307323391","https://openalex.org/W4372191700","https://openalex.org/W4372349107","https://openalex.org/W4385570550","https://openalex.org/W4385823270","https://openalex.org/W4385823331","https://openalex.org/W4385893869","https://openalex.org/W4389600306","https://openalex.org/W4392903062","https://openalex.org/W6631190155","https://openalex.org/W6760633627","https://openalex.org/W6780218876","https://openalex.org/W6785521828","https://openalex.org/W6838789019","https://openalex.org/W6840426711","https://openalex.org/W6841035593","https://openalex.org/W6852720755","https://openalex.org/W6853532542","https://openalex.org/W6853998256","https://openalex.org/W6855650468"],"related_works":["https://openalex.org/W2883671469","https://openalex.org/W2728761353","https://openalex.org/W2338806053","https://openalex.org/W4385571610","https://openalex.org/W4397292310","https://openalex.org/W39235475","https://openalex.org/W1540615732","https://openalex.org/W2500421879","https://openalex.org/W2537969829","https://openalex.org/W29040644"],"abstract_inverted_index":{"Simultaneous":[0],"speech-to-speech":[1],"translation":[2,22],"(S2ST)":[3],"holds":[4],"the":[5,51,54,88,109],"promise":[6],"of":[7,53],"breaking":[8],"down":[9],"communication":[10],"barriers":[11],"and":[12,64,100],"enabling":[13],"fluid":[14],"conversations":[15],"across":[16],"languages.":[17],"However,":[18],"achieving":[19],"accurate,":[20],"real-time":[21,81,105],"through":[23],"mobile":[24],"devices":[25],"remains":[26],"a":[27,33,44,119],"major":[28],"challenge.":[29],"We":[30],"introduce":[31],"SimulTron,":[32],"novel":[34],"S2ST":[35,47,106,129],"architecture":[36],"designed":[37],"to":[38,103],"tackle":[39],"this":[40],"task.":[41],"SimulTron":[42,73,85,95,117],"is":[43],"lightweight":[45],"direct":[46],"model":[48],"that":[49,72,84],"uses":[50],"strengths":[52],"Translatotron":[55,75,92],"framework":[56],"while":[57],"incorporating":[58],"key":[59],"modifications":[60],"for":[61,127],"streaming":[62],"operation,":[63],"an":[65],"adjustable":[66],"fixed":[67],"delay.":[68],"Our":[69],"experiments":[70],"show":[71,124],"surpasses":[74],"2":[76],"in":[77],"offline":[78],"evaluations.":[79],"Furthermore,":[80],"evaluations":[82],"reveal":[83],"improves":[86],"upon":[87],"performance":[89],"achieved":[90],"by":[91],"1.":[93],"Additionally,":[94],"achieves":[96],"superior":[97],"BLEU":[98],"scores":[99],"latency":[101],"compared":[102],"previous":[104],"method":[107],"on":[108,118],"MuST-C":[110],"dataset.":[111],"Significantly,":[112],"we":[113],"have":[114],"successfully":[115],"deployed":[116],"Pixel":[120],"7":[121],"Pro":[122],"device,":[123],"its":[125],"potential":[126],"simultaneous":[128],"on-device.":[130]},"counts_by_year":[{"year":2025,"cited_by_count":2}],"updated_date":"2025-12-21T01:58:51.020947","created_date":"2025-10-10T00:00:00"}