{"id":"https://openalex.org/W7225388270","doi":"https://doi.org/10.48550/arxiv.2611.06982","title":"Decode Before You Hear: Speculative Listening for Chunk-Interleaved Full-Duplex Speech Language Models","display_name":"Decode Before You Hear: Speculative Listening for Chunk-Interleaved Full-Duplex Speech Language Models","publication_year":2026,"publication_date":"2026-11-04","ids":{"openalex":"https://openalex.org/W7225388270","doi":"https://doi.org/10.48550/arxiv.2611.06982"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2611.06982","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2611.06982","pdf_url":"https://arxiv.org/pdf/2611.06982","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","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2611.06982","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5138285542","display_name":"Yutong Bian","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bian, Yutong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5161042616","display_name":"Yanglin Zhang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Yanglin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5161072258","display_name":"Yuxiang Wang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Yuxiang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134958356","display_name":"Luoyuan Zhang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Luoyuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5112458015","display_name":"Rong\u2010De He","orcid":"https://orcid.org/0000-0003-0397-5367","observed_orcids":["https://orcid.org/0000-0003-0397-5367"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"He, Rongshen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5161080454","display_name":"Zihao Wang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Zihao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5161024015","display_name":"Junwen Qiu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qiu, Junwen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5161126970","display_name":"Jiaqi Li","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Jiaqi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5161020102","display_name":"Zhizheng Wu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Zhizheng","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":0.8134999871253967,"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":0.8134999871253967,"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":0.1177000030875206,"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":0.030799999833106995,"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/speculative-decoding","display_name":"speculative decoding","score":0.800000011920929},{"id":"https://openalex.org/keywords/speech-language-models","display_name":"speech language models","score":0.6859999895095825},{"id":"https://openalex.org/keywords/decoding-delay-reduction","display_name":"decoding delay reduction","score":0.5799999833106995},{"id":"https://openalex.org/keywords/tool-calls","display_name":"tool calls","score":0.6129999756813049}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7982000112533569},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.7020000219345093},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.60589998960495},{"id":"https://openalex.org/C2781115785","wikidata":"https://www.wikidata.org/wiki/Q502261","display_name":"Silence","level":2,"score":0.5819000005722046},{"id":"https://openalex.org/C177291462","wikidata":"https://www.wikidata.org/wiki/Q423038","display_name":"Active listening","level":2,"score":0.5644999742507935},{"id":"https://openalex.org/C2777200299","wikidata":"https://www.wikidata.org/wiki/Q52943","display_name":"Conversation","level":2,"score":0.527400016784668},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.51419997215271},{"id":"https://openalex.org/C81081738","wikidata":"https://www.wikidata.org/wiki/Q55542","display_name":"Lossless compression","level":3,"score":0.5067999958992004},{"id":"https://openalex.org/C75778745","wikidata":"https://www.wikidata.org/wiki/Q342626","display_name":"Lag","level":2,"score":0.4311999976634979},{"id":"https://openalex.org/C16320812","wikidata":"https://www.wikidata.org/wiki/Q1812200","display_name":"Idle","level":2,"score":0.4174000024795532},{"id":"https://openalex.org/C2778476105","wikidata":"https://www.wikidata.org/wiki/Q628539","display_name":"Workload","level":2,"score":0.398499995470047},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.367000013589859},{"id":"https://openalex.org/C13895895","wikidata":"https://www.wikidata.org/wiki/Q3270773","display_name":"Speech coding","level":2,"score":0.3037000000476837},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.30079999566078186},{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.29109999537467957},{"id":"https://openalex.org/C2776141515","wikidata":"https://www.wikidata.org/wiki/Q1274479","display_name":"Repetition (rhetorical device)","level":2,"score":0.27570000290870667},{"id":"https://openalex.org/C43617652","wikidata":"https://www.wikidata.org/wiki/Q7575399","display_name":"Speech production","level":2,"score":0.27090001106262207},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.258899986743927},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.2565999925136566},{"id":"https://openalex.org/C2778915421","wikidata":"https://www.wikidata.org/wiki/Q3643177","display_name":"Performance improvement","level":2,"score":0.25519999861717224}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2611.06982","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2611.06982","pdf_url":"https://arxiv.org/pdf/2611.06982","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.2611.06982","doi":"https://doi.org/10.48550/arxiv.2611.06982","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2611.06982","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:2611.06982","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2611.06982","pdf_url":"https://arxiv.org/pdf/2611.06982","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":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7225388270.pdf","grobid_xml":"https://content.openalex.org/works/W7225388270.grobid-xml"},"referenced_works_count":19,"referenced_works":["https://openalex.org/W2612431505","https://openalex.org/W4281690148","https://openalex.org/W4319166707","https://openalex.org/W4402672007","https://openalex.org/W4402683901","https://openalex.org/W4403883071","https://openalex.org/W4404783093","https://openalex.org/W4412945230","https://openalex.org/W4412945522","https://openalex.org/W4414690747","https://openalex.org/W4414849600","https://openalex.org/W4414973504","https://openalex.org/W4415334162","https://openalex.org/W7140282780","https://openalex.org/W7148492801","https://openalex.org/W7151544009","https://openalex.org/W7166845265","https://openalex.org/W7213284145","https://openalex.org/W7214115718"],"related_works":["https://openalex.org/W4415334162","https://openalex.org/W7203483839","https://openalex.org/W7221757954","https://openalex.org/W7168078438","https://openalex.org/W7171963079","https://openalex.org/W4402112402","https://openalex.org/W7169850532","https://openalex.org/W7135233266","https://openalex.org/W4403346564","https://openalex.org/W7139955336"],"abstract_inverted_index":{"Chunk-interleaved":[0],"full-duplex":[1],"speech":[2],"language":[3],"models":[4],"listen":[5,29],"and":[6,12,37,39,48,76,168,188,220],"speak":[7,38,184],"in":[8,103,163],"fixed":[9],"time":[10,226],"chunks,":[11],"each":[13],"chunk":[14,58,97,141],"must":[15],"be":[16],"processed":[17],"before":[18,62,202],"the":[19,52,56,73,78,81,94,101,113,136,139,148,155,172,199,213,221],"next":[20,57,95,140],"one":[21],"arrives.":[22,174],"Their":[23],"workload":[24],"is":[25],"asymmetric":[26],"across":[27],"chunks:":[28],"chunks":[30,41,185,190],"leave":[31],"compute":[32,54],"idle":[33,53,134],"while":[34],"awaiting":[35],"audio,":[36],"tool-call":[40,189],"require":[42],"serial":[43],"generation,":[44],"which":[45],"delays":[46],"responses":[47],"tool":[49],"calls.":[50],"Using":[51],"for":[55,129,145],"requires":[59],"decoding":[60,200],"it":[61,197],"its":[63],"audio":[64,85,96,173],"arrives,":[65],"under":[66,83,142],"an":[67],"assumed":[68,74],"input.":[69],"Taking":[70],"silence":[71,143],"as":[72],"input":[75],"comparing":[77],"output":[79,82,102,115],"with":[80],"real":[84,225],"on":[86,118],"four":[87],"models,":[88],"we":[89,121],"observe":[90],"two":[91],"properties.":[92],"First,":[93],"does":[98],"not":[99],"change":[100],"95.3%":[104],"of":[105,109],"chunks.":[106],"Second,":[107],"88%":[108],"changes":[110,146],"begin":[111],"at":[112,147],"first":[114,149,157],"token.":[116,158],"Building":[117],"these":[119],"properties,":[120],"propose":[122],"Speculative":[123,180],"Listening,":[124],"a":[125,151,164,203],"training-free,":[126],"lossless":[127],"framework":[128],"input-time":[130],"self-speculative":[131],"decoding.":[132],"During":[133],"time,":[135],"model":[137],"drafts":[138,162],"and,":[144],"token,":[150],"second":[152],"branch":[153],"from":[154],"runner-up":[156],"It":[159],"keeps":[160],"all":[161],"cross-chunk":[165],"hypothesis":[166],"pool":[167],"verifies":[169],"them":[170],"when":[171],"On":[175,195],"our":[176],"eight-task":[177],"benchmark":[178],"TickTock-Bench,":[179],"Listening":[181],"speeds":[182],"up":[183,192],"by":[186,191,206,227],"2.6-2.8x":[187],"to":[193,208,212],"23.4x.":[194],"average,":[196],"reduces":[198],"delay":[201],"response":[204],"starts":[205],"73%,":[207],"74-241":[209],"ms,":[210],"close":[211],"200":[214],"ms":[215],"gap":[216],"between":[217],"human":[218],"turns,":[219],"maximum":[222],"lag":[223],"behind":[224],"94.4%.":[228]},"counts_by_year":[],"updated_date":"2026-11-09T13:22:03.289127","created_date":"2026-11-07T00:00:00"}