{"id":"https://openalex.org/W7213242585","doi":"https://doi.org/10.48550/arxiv.2609.14991","title":"Typhoon ASR Streaming: Steerable Low-Latency Thai Speech Recognition with Real-Time Shallow Fusion","display_name":"Typhoon ASR Streaming: Steerable Low-Latency Thai Speech Recognition with Real-Time Shallow Fusion","publication_year":2026,"publication_date":"2026-09-14","ids":{"openalex":"https://openalex.org/W7213242585","doi":"https://doi.org/10.48550/arxiv.2609.14991"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2609.14991","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.14991","pdf_url":"https://arxiv.org/pdf/2609.14991","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/2609.14991","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5114387463","display_name":"Warit Sirichotedumrong","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sirichotedumrong, Warit","raw_affiliation_strings":["Typhoon Team , SCB DataX , Bangkok , Thailand"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Typhoon Team , SCB DataX , Bangkok , Thailand","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5151710458","display_name":"Tanawin Samutsin","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Samutsin, Tanawin","raw_affiliation_strings":["Typhoon Team , SCB DataX , Bangkok , Thailand"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Typhoon Team , SCB DataX , Bangkok , Thailand","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5151755038","display_name":"Shah Faisal Wani","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wani, Shah Faisal","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5093567857","display_name":"Sittipong Sripaisarnmongkol","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sripaisarnmongkol, Sittipong","raw_affiliation_strings":["Typhoon Team , SCB DataX , Bangkok , Thailand"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Typhoon Team , SCB DataX , Bangkok , Thailand","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5021237405","display_name":"Kunat Pipatanakul","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pipatanakul, Kunat","raw_affiliation_strings":["Typhoon Team , SCB DataX , Bangkok , Thailand"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Typhoon Team , SCB DataX , Bangkok , Thailand","institution_ids":[]}]}],"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":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":[{"id":"https://openalex.org/keywords/thai-speech-recognition","display_name":"Thai speech recognition","score":0.8360000252723694},{"id":"https://openalex.org/keywords/streaming-asr","display_name":"streaming ASR","score":0.9229999780654907},{"id":"https://openalex.org/keywords/low-latency-decoding","display_name":"low-latency decoding","score":0.8109999895095825},{"id":"https://openalex.org/keywords/shallow-fusion","display_name":"shallow fusion","score":0.8109999895095825},{"id":"https://openalex.org/keywords/n-gram-language-models","display_name":"n-gram language models","score":0.8939999938011169},{"id":"https://openalex.org/keywords/code-switching","display_name":"code-switching","score":0.7559999823570251},{"id":"https://openalex.org/keywords/live-captioning","display_name":"live captioning","score":0.7269999980926514},{"id":"https://openalex.org/keywords/voice-assistants","display_name":"voice assistants","score":0.7099999785423279}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8080999851226807},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.728600025177002},{"id":"https://openalex.org/C2776224158","wikidata":"https://www.wikidata.org/wiki/Q187931","display_name":"Phrase","level":2,"score":0.6029999852180481},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.46630001068115234},{"id":"https://openalex.org/C46686674","wikidata":"https://www.wikidata.org/wiki/Q466303","display_name":"Boosting (machine learning)","level":2,"score":0.4636000096797943},{"id":"https://openalex.org/C2780615836","wikidata":"https://www.wikidata.org/wiki/Q2471869","display_name":"USable","level":2,"score":0.45660001039505005},{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.4422999918460846},{"id":"https://openalex.org/C2777601683","wikidata":"https://www.wikidata.org/wiki/Q6499736","display_name":"Vocabulary","level":2,"score":0.4401000142097473},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.41530001163482666},{"id":"https://openalex.org/C2775852435","wikidata":"https://www.wikidata.org/wiki/Q258403","display_name":"Utterance","level":2,"score":0.41110000014305115},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.3971000015735626},{"id":"https://openalex.org/C23224414","wikidata":"https://www.wikidata.org/wiki/Q176769","display_name":"Hidden Markov model","level":2,"score":0.3837999999523163},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.3596999943256378},{"id":"https://openalex.org/C127759330","wikidata":"https://www.wikidata.org/wiki/Q637416","display_name":"Codebook","level":2,"score":0.3271999955177307},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3073999881744385},{"id":"https://openalex.org/C2779119184","wikidata":"https://www.wikidata.org/wiki/Q294350","display_name":"ALARM","level":2,"score":0.2732999920845032},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.2694999873638153},{"id":"https://openalex.org/C141603448","wikidata":"https://www.wikidata.org/wiki/Q134830","display_name":"Prefix","level":2,"score":0.26820001006126404},{"id":"https://openalex.org/C117884012","wikidata":"https://www.wikidata.org/wiki/Q94489","display_name":"n-gram","level":3,"score":0.263700008392334},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.26109999418258667},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.25920000672340393},{"id":"https://openalex.org/C2780945871","wikidata":"https://www.wikidata.org/wiki/Q194274","display_name":"Backup","level":2,"score":0.25870001316070557}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2609.14991","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.14991","pdf_url":"https://arxiv.org/pdf/2609.14991","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.2609.14991","doi":"https://doi.org/10.48550/arxiv.2609.14991","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2609.14991","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:2609.14991","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.14991","pdf_url":"https://arxiv.org/pdf/2609.14991","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/W7213242585.pdf","grobid_xml":"https://content.openalex.org/works/W7213242585.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7125200645","https://openalex.org/W7135169363","https://openalex.org/W7076811890","https://openalex.org/W4206017443","https://openalex.org/W7212059113","https://openalex.org/W7128812861","https://openalex.org/W4385570104","https://openalex.org/W7127203856","https://openalex.org/W3157408717","https://openalex.org/W7164144082"],"abstract_inverted_index":{"Open":[0],"Thai":[1,37,55,107],"automatic":[2],"speech":[3],"recognition":[4],"(ASR)":[5],"is":[6],"dominated":[7],"by":[8,75],"offline,":[9],"Whisper-based":[10],"models":[11,115],"that":[12,39,89],"read":[13],"the":[14,93,113,119,157,169,175],"whole":[15],"utterance":[16],"before":[17],"transcribing,":[18],"ruling":[19],"out":[20],"low-latency":[21],"uses":[22],"such":[23],"as":[24,64],"live":[25],"captioning":[26],"and":[27,84,102,109,152],"voice":[28],"agents.":[29],"We":[30],"present":[31],"a":[32,41,65,72,80,86,97,129],"deployable":[33],"system":[34],"for":[35],"streaming":[36,70,82,94,114],"ASR":[38],"lets":[40],"user":[42],"steer":[43],"its":[44],"vocabulary":[45],"at":[46,128,148],"decode":[47],"time,":[48],"without":[49],"retraining.":[50],"A":[51],"widely":[52],"used":[53],"open":[54],"model,":[56],"trained":[57],"with":[58,71,96,181],"full":[59],"context,":[60],"collapses":[61],"when":[62],"run":[63],"true":[66],"stream;":[67],"we":[68],"restore":[69],"cache-aware":[73],"encoder,":[74],"converting":[76],"it":[77],"or":[78],"adapting":[79],"natively":[81],"one,":[83],"add":[85],"shallow-fusion":[87],"layer":[88],"re-ranks":[90],"candidates":[91],"inside":[92],"decoder":[95],"GPU":[98],"n-gram":[99,162],"language":[100],"model":[101,111,121,176],"phrase":[103,182],"boosting.":[104],"Across":[105],"two":[106,110],"benchmarks":[108],"sizes,":[112],"stay":[116],"usable":[117],"where":[118],"full-context":[120],"fails,":[122],"cutting":[123],"character":[124],"error":[125],"rate":[126],"4.3-4.5x":[127],"one-second":[130],"look-ahead":[131],"while":[132],"running":[133],"faster":[134],"than":[135],"real":[136],"time.":[137],"Decode-time":[138],"steering":[139],"then":[140],"lifts":[141],"keyword":[142],"recall":[143],"from":[144,160],"16.6%":[145],"to":[146],"20.7%":[147],"no":[149],"accuracy":[150],"cost":[151],"negligible":[153],"overhead;":[154],"most":[155],"of":[156,172],"gain":[158],"comes":[159],"an":[161],"over":[163,187],"ordinary":[164],"training":[165],"transcripts,":[166],"which":[167],"resolves":[168],"written":[170],"form":[171],"code-switched":[173],"words":[174],"hears":[177],"but":[178],"spells":[179],"inconsistently,":[180],"boosting":[183],"adding":[184],"targeted":[185],"control":[186],"rare":[188],"domain":[189],"terms.":[190]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-09-16T00:00:00"}