{"id":"https://openalex.org/W7225833358","doi":"https://doi.org/10.48550/arxiv.2611.14421","title":"Earlybird: Predictive Streaming Speech Recognition in 29M Parameters","display_name":"Earlybird: Predictive Streaming Speech Recognition in 29M Parameters","publication_year":2026,"publication_date":"2026-11-10","ids":{"openalex":"https://openalex.org/W7225833358","doi":"https://doi.org/10.48550/arxiv.2611.14421"},"language":"en","primary_location":{"id":"doi:10.48550/arxiv.2611.14421","doi":"https://doi.org/10.48550/arxiv.2611.14421","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2611.14421","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2611.14421","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5157777358","display_name":"Sahil Mahendrakar","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Mahendrakar, Sahil","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5157777358"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"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":1.0,"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":1.0,"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":9.999999747378752e-05,"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/T12031","display_name":"Speech and dialogue systems","score":0.0,"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":[],"concepts":[{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.7588000297546387},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.7466999888420105},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7427999973297119},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.6990000009536743},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5188000202178955},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3912999927997589},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3479999899864197},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.3370000123977661},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.3228999972343445},{"id":"https://openalex.org/C169903167","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Test set","level":2,"score":0.32030001282691956},{"id":"https://openalex.org/C2778067643","wikidata":"https://www.wikidata.org/wiki/Q166507","display_name":"Interval (graph theory)","level":2,"score":0.319599986076355},{"id":"https://openalex.org/C196083921","wikidata":"https://www.wikidata.org/wiki/Q7915758","display_name":"Variance (accounting)","level":2,"score":0.31679999828338623},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.2913999855518341},{"id":"https://openalex.org/C204201278","wikidata":"https://www.wikidata.org/wiki/Q1332614","display_name":"Voice activity detection","level":3,"score":0.27639999985694885},{"id":"https://openalex.org/C157968479","wikidata":"https://www.wikidata.org/wiki/Q3079876","display_name":"Audio mining","level":4,"score":0.2703999876976013},{"id":"https://openalex.org/C150856459","wikidata":"https://www.wikidata.org/wiki/Q8034367","display_name":"Word recognition","level":3,"score":0.2515000104904175}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2611.14421","doi":"https://doi.org/10.48550/arxiv.2611.14421","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2611.14421","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2611.14421","doi":"https://doi.org/10.48550/arxiv.2611.14421","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2611.14421","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Streaming":[0],"speech":[1,109],"recognizers":[2],"output":[3],"a":[4,18,36,45,66,83,95,219],"word":[5,74,99,114],"only":[6],"after":[7,135],"the":[8,24,26,30,32,86,101,112,121,140,161,176,198,222],"speaker":[9,122],"has":[10],"finished":[11],"it,":[12,124],"typically":[13],"90-400":[14],"ms":[15,119,134],"later.":[16],"For":[17,179],"voice":[19],"agent":[20,33],"every":[21,73,98,213],"millisecond":[22],"counts:":[23],"sooner":[25,31],"words":[27,166,180],"are":[28,167],"outputted,":[29],"can":[34],"start":[35],"search":[37],"or":[38],"plan":[39],"its":[40],"reply.":[41],"We":[42],"present":[43],"Earlybird,":[44],"29M-parameter":[46],"streaming":[47,127],"recognizer":[48],"(16.8":[49],"MB":[50],"at":[51,60],"4":[52],"bits).":[53],"Earlybird":[54,138],"learns":[55],"from":[56],"audio":[57],"cut":[58],"off":[59],"random":[61],"points,":[62],"including":[63,100],"partway":[64],"through":[65],"word,":[67],"and":[68,157,215],"is":[69,76,115,139,185,206],"asked":[70],"to":[71,132,195],"predict":[72],"that":[75],"due":[77],"by":[78],"then":[79],"in":[80],"full,":[81],"plus":[82],"guess":[84],"of":[85,165,193,221],"next":[87],"two":[88],"words.":[89],"Its":[90],"raw":[91],"token":[92],"probabilities":[93],"give":[94],"confidence":[96,184],"for":[97,175],"one":[102],"still":[103,181],"being":[104,182],"spoken.":[105],"On":[106,160,197],"clean":[107],"audiobook":[108],"(LibriSpeech":[110],"test-clean),":[111],"median":[113],"predicted":[116,168],"correctly":[117,169],"90":[118,131],"before":[120,170],"finishes":[123],"while":[125],"released":[126],"models":[128],"get":[129],"it":[130,136,205],"400":[133],"ends.":[137],"earliest":[141],"on":[142,212],"all":[143],"six":[144],"test":[145],"sets":[146],"we":[147],"time,":[148],"covering":[149],"audiobooks,":[150,163],"TED":[151],"talks,":[152],"public":[153],"speech,":[154],"earnings":[155],"calls,":[156],"synthetic":[158],"speech.":[159],"same":[162,223],"69%":[164],"they":[171],"end,":[172],"against":[173],"3-25%":[174],"other":[177],"models.":[178],"spoken,":[183],"calibrated,":[186],"with":[187],"an":[188],"expected":[189],"calibration":[190],"error":[191],"(ECE)":[192],"3.7":[194],"5.5.":[196],"eight":[199],"English":[200],"Open":[201],"ASR":[202],"Leaderboard":[203],"sets,":[204],"more":[207],"accurate":[208],"than":[209],"Moonshine":[210],"Tiny":[211],"set":[214],"roughly":[216],"ties":[217],"Whistle,":[218],"model":[220],"size.":[224]},"counts_by_year":[],"updated_date":"2026-11-12T12:28:57.059296","created_date":"2026-11-12T00:00:00"}