{"id":"https://openalex.org/W7213611583","doi":"https://doi.org/10.48550/arxiv.2609.19334","title":"A frontend-backend architecture for tool calls in full-duplex speech models","display_name":"A frontend-backend architecture for tool calls in full-duplex speech models","publication_year":2026,"publication_date":"2026-09-16","ids":{"openalex":"https://openalex.org/W7213611583","doi":"https://doi.org/10.48550/arxiv.2609.19334"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2609.19334","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.19334","pdf_url":"https://arxiv.org/pdf/2609.19334","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2609.19334","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5151760937","display_name":"Ke Hu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hu, Ke","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5083696320","display_name":"Slyne Deng","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Deng, Slyne","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5152139998","display_name":"Chen Chen","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Chen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5007900374","display_name":"Elena Rastorgueva","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Rastorgueva, Elena","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5025318789","display_name":"Edresson Casanova","orcid":"https://orcid.org/0000-0003-0160-7173","observed_orcids":["https://orcid.org/0000-0003-0160-7173"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Casanova, Edresson","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5152112213","display_name":"Punit Kumar","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kumar, Punit","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5152132044","display_name":"Dharmendra Choudhary","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Choudhary, Dharmendra","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137358714","display_name":"Nikhil Srihari","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Srihari, Nikhil","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063510148","display_name":"Ameya Sunil Mahabaleshwarkar","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mahabaleshwarkar, Ameya Sunil","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5085746637","display_name":"Viet Anh Trinh","orcid":"https://orcid.org/0000-0002-1660-6627","observed_orcids":["https://orcid.org/0000-0002-1660-6627"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Trinh, Viet Anh","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060031161","display_name":"Slim Essid","orcid":"https://orcid.org/0000-0002-0028-327X","observed_orcids":["https://orcid.org/0000-0002-0028-327X"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Essid, Slim","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5124283176","display_name":"Oluwatobi Olabiyi","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Olabiyi, Oluwatobi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5002433660","display_name":"Zhehuai Chen","orcid":"https://orcid.org/0000-0003-4400-5340","observed_orcids":["https://orcid.org/0000-0003-4400-5340"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Zhehuai","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.7598000168800354,"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":0.7598000168800354,"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":0.23829999566078186,"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":0.0010000000474974513,"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/tool-calling","display_name":"tool calling","score":0.8240000009536743},{"id":"https://openalex.org/keywords/streaming-asr","display_name":"streaming ASR","score":0.6499999761581421},{"id":"https://openalex.org/keywords/voice-assistants","display_name":"voice assistants","score":0.7239999771118164},{"id":"https://openalex.org/keywords/turn-taking","display_name":"turn-taking","score":0.6269999742507935},{"id":"https://openalex.org/keywords/agentic-llms","display_name":"agentic LLMs","score":0.5180000066757202}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7788000106811523},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.7063999772071838},{"id":"https://openalex.org/C101468663","wikidata":"https://www.wikidata.org/wiki/Q1620158","display_name":"Modular design","level":2,"score":0.6078000068664551},{"id":"https://openalex.org/C99611785","wikidata":"https://www.wikidata.org/wiki/Q6453233","display_name":"Duplex (building)","level":3,"score":0.5454999804496765},{"id":"https://openalex.org/C86532276","wikidata":"https://www.wikidata.org/wiki/Q1184065","display_name":"Delegation","level":2,"score":0.48410001397132874},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.44699999690055847},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.4056999981403351},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.37689998745918274},{"id":"https://openalex.org/C98025372","wikidata":"https://www.wikidata.org/wiki/Q477538","display_name":"Systems architecture","level":3,"score":0.36149999499320984},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.3416000008583069},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.33640000224113464},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.29739999771118164},{"id":"https://openalex.org/C89611455","wikidata":"https://www.wikidata.org/wiki/Q6804646","display_name":"Mechanism (biology)","level":2,"score":0.26989999413490295},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.2662000060081482},{"id":"https://openalex.org/C14999030","wikidata":"https://www.wikidata.org/wiki/Q16346","display_name":"Speech synthesis","level":2,"score":0.26409998536109924},{"id":"https://openalex.org/C190954187","wikidata":"https://www.wikidata.org/wiki/Q5270587","display_name":"Dialog system","level":3,"score":0.2572000026702881}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2609.19334","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.19334","pdf_url":"https://arxiv.org/pdf/2609.19334","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2609.19334","doi":"https://doi.org/10.48550/arxiv.2609.19334","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2609.19334","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2609.19334","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.19334","pdf_url":"https://arxiv.org/pdf/2609.19334","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7213611583.pdf","grobid_xml":"https://content.openalex.org/works/W7213611583.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7213936344","https://openalex.org/W7161968102","https://openalex.org/W4417262065","https://openalex.org/W7214788875","https://openalex.org/W7212080144","https://openalex.org/W7164829614","https://openalex.org/W7156565615","https://openalex.org/W7148277713","https://openalex.org/W7213312795","https://openalex.org/W7155177999"],"abstract_inverted_index":{"Full-duplex":[0],"speech-to-speech":[1],"(S2S)":[2],"models":[3],"provide":[4],"natural,":[5],"low-latency":[6,87],"conversational":[7],"interaction":[8,88,167],"and":[9,19,39,68,86,113,138,142,146,159],"would":[10],"benefit":[11],"from":[12,54],"the":[13,55,61,75,95],"ability":[14],"to":[15,34,44,74,94,136],"use":[16],"external":[17],"tools":[18],"complete":[20],"voice-agent":[21],"tasks.":[22],"We":[23],"propose":[24],"a":[25,29,36,45,64,99,123],"frontend-backend":[26],"architecture":[27],"where":[28],"duplex":[30,82,165],"speech-to-text":[31],"frontend":[32,62,96],"learns":[33],"emit":[35],"delegation":[37,155],"token":[38],"forwards":[40],"streaming":[41,72],"ASR":[42],"transcripts":[43],"text-based":[46],"backend":[47,56,125,154],"LLM":[48],"for":[49,162],"tool":[50],"calls.":[51,119],"Tool-call":[52],"results":[53,132,151],"are":[57],"injected":[58],"back":[59],"into":[60],"through":[63],"lightweight":[65],"prefill-and-repeat":[66],"mechanism":[67],"then":[69],"synthesized":[70],"using":[71],"TTS":[73],"user.":[76],"Our":[77],"approach":[78,161],"largely":[79],"preserves":[80],"regular":[81],"turn-taking,":[83],"interruption":[84],"handling,":[85],"as":[89],"it":[90],"requires":[91],"minimal":[92],"modifications":[93],"model.":[97],"In":[98],"single-turn":[100],"tool-call":[101,107,110,171],"evaluation,":[102],"our":[103,128],"system":[104,129],"achieves":[105,130],"92-97%":[106],"recall,":[108],"competitive":[109,131],"prediction":[111],"performance,":[112],"81.2%":[114],"accuracy":[115],"in":[116],"rejecting":[117],"irrelevant":[118],"When":[120],"equipped":[121],"with":[122,168],"larger":[124],"(e.g.,":[126],"Qwen3-235B-A22B),":[127],"on":[133,148],"Full-Duplex-Bench-V3":[134],"compared":[135],"open":[137],"closed":[139],"source":[140],"models,":[141],"significantly":[143],"outperforms":[144],"GPT-realtime-mini":[145],"Qwen3-Omni-30B-A3B-Instruct":[147],"EVA-Bench.":[149],"These":[150],"demonstrate":[152],"that":[153],"is":[156],"an":[157],"effective":[158],"modular":[160],"combining":[163],"natural":[164],"speech":[166],"strong":[169],"agentic":[170],"capabilities.":[172]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-09-19T00:00:00"}