{"id":"https://openalex.org/W4400070537","doi":"https://doi.org/10.1109/lsp.2024.3419719","title":"Tuning Large Language Model for Speech Recognition With Mixed-Scale Re-Tokenization","display_name":"Tuning Large Language Model for Speech Recognition With Mixed-Scale Re-Tokenization","publication_year":2024,"publication_date":"2024-01-01","ids":{"openalex":"https://openalex.org/W4400070537","doi":"https://doi.org/10.1109/lsp.2024.3419719"},"language":"en","primary_location":{"id":"doi:10.1109/lsp.2024.3419719","is_oa":false,"landing_page_url":"https://doi.org/10.1109/lsp.2024.3419719","pdf_url":null,"source":{"id":"https://openalex.org/S120629676","display_name":"IEEE Signal Processing Letters","issn_l":"1070-9908","issn":["1070-9908","1558-2361"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Signal Processing Letters","raw_type":"journal-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/A5025199890","display_name":"Yukun Ma","orcid":"https://orcid.org/0000-0002-0520-4331"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yukun Ma","raw_affiliation_strings":["Speech Lab, Institute for Intelligent Computing, Alibaba Group, Singapore"],"raw_orcid":"https://orcid.org/0000-0002-0520-4331","affiliations":[{"raw_affiliation_string":"Speech Lab, Institute for Intelligent Computing, Alibaba Group, Singapore","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100393899","display_name":"Chong Zhang","orcid":"https://orcid.org/0000-0002-2162-4344"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chong Zhang","raw_affiliation_strings":["Speech Lab, Institute for Intelligent Computing, Alibaba Group, Singapore"],"raw_orcid":"https://orcid.org/0000-0002-2162-4344","affiliations":[{"raw_affiliation_string":"Speech Lab, Institute for Intelligent Computing, Alibaba Group, Singapore","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100428431","display_name":"Qian Chen","orcid":"https://orcid.org/0000-0001-6939-7438"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qian Chen","raw_affiliation_strings":["Speech Lab, Institute for Intelligent Computing, Alibaba Group, Singapore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Speech Lab, Institute for Intelligent Computing, Alibaba Group, Singapore","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100396375","display_name":"Wen Wang","orcid":"https://orcid.org/0000-0002-0356-1968"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wen Wang","raw_affiliation_strings":["Speech Lab, Institute for Intelligent Computing, Alibaba Group, Singapore"],"raw_orcid":"https://orcid.org/0000-0002-0356-1968","affiliations":[{"raw_affiliation_string":"Speech Lab, Institute for Intelligent Computing, Alibaba Group, Singapore","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100595427","display_name":"Bin Ma","orcid":"https://orcid.org/0000-0002-3724-4581"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bin Ma","raw_affiliation_strings":["Speech Lab, Institute for Intelligent Computing, Alibaba Group, Singapore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Speech Lab, Institute for Intelligent Computing, Alibaba Group, Singapore","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.6671,"has_fulltext":false,"cited_by_count":6,"citation_normalized_percentile":{"value":0.85901986,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":98},"biblio":{"volume":"31","issue":null,"first_page":"1740","last_page":"1744"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9929999709129333,"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":0.9929999709129333,"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":0.9384999871253967,"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/computer-science","display_name":"Computer science","score":0.7958341836929321},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.642138659954071},{"id":"https://openalex.org/keywords/lexical-analysis","display_name":"Lexical analysis","score":0.6187947988510132},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.547105610370636},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4976842701435089},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.4940991699695587},{"id":"https://openalex.org/keywords/scale","display_name":"Scale (ratio)","score":0.4905985891819},{"id":"https://openalex.org/keywords/hidden-markov-model","display_name":"Hidden Markov model","score":0.42022013664245605}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7958341836929321},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.642138659954071},{"id":"https://openalex.org/C176982825","wikidata":"https://www.wikidata.org/wiki/Q835922","display_name":"Lexical analysis","level":2,"score":0.6187947988510132},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.547105610370636},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4976842701435089},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4940991699695587},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.4905985891819},{"id":"https://openalex.org/C23224414","wikidata":"https://www.wikidata.org/wiki/Q176769","display_name":"Hidden Markov model","level":2,"score":0.42022013664245605},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/lsp.2024.3419719","is_oa":false,"landing_page_url":"https://doi.org/10.1109/lsp.2024.3419719","pdf_url":null,"source":{"id":"https://openalex.org/S120629676","display_name":"IEEE Signal Processing Letters","issn_l":"1070-9908","issn":["1070-9908","1558-2361"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Signal Processing Letters","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":41,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W2963250244","https://openalex.org/W3180374548","https://openalex.org/W3209059054","https://openalex.org/W3209873929","https://openalex.org/W4226507725","https://openalex.org/W4292825791","https://openalex.org/W4322718191","https://openalex.org/W4378501656","https://openalex.org/W4378711593","https://openalex.org/W4381786045","https://openalex.org/W4385328213","https://openalex.org/W4385571769","https://openalex.org/W4385573012","https://openalex.org/W4385573788","https://openalex.org/W4385805202","https://openalex.org/W4385822337","https://openalex.org/W4385822683","https://openalex.org/W4386351448","https://openalex.org/W4389520395","https://openalex.org/W4389524500","https://openalex.org/W4390190613","https://openalex.org/W4391021623","https://openalex.org/W4391021666","https://openalex.org/W4391021773","https://openalex.org/W4391021779","https://openalex.org/W4391835492","https://openalex.org/W4392902656","https://openalex.org/W6796581206","https://openalex.org/W6797019122","https://openalex.org/W6800875267","https://openalex.org/W6850625674","https://openalex.org/W6852447913","https://openalex.org/W6853611000","https://openalex.org/W6855691466","https://openalex.org/W6855801281","https://openalex.org/W6855885476","https://openalex.org/W6856179682","https://openalex.org/W6857054612","https://openalex.org/W6860809563","https://openalex.org/W6861626678"],"related_works":["https://openalex.org/W2136763963","https://openalex.org/W2109705048","https://openalex.org/W2940588515","https://openalex.org/W1909151225","https://openalex.org/W1987783679","https://openalex.org/W2160030256","https://openalex.org/W1521297879","https://openalex.org/W4253235840","https://openalex.org/W3151937861","https://openalex.org/W4300598845"],"abstract_inverted_index":{"Large":[0],"Language":[1],"Models":[2],"(LLMs)":[3],"have":[4],"proven":[5],"successful":[6],"across":[7],"a":[8,91,136],"spectrum":[9],"of":[10,25,55,66,121,126,129,135,144],"speech-related":[11],"tasks,":[12],"such":[13],"as":[14,32],"speech":[15,27,69,81,100,146],"recognition,":[16],"text-to-speech,":[17],"and":[18,35,52,82],"spoken":[19],"language":[20],"understanding.":[21],"Recently,":[22],"the":[23,63,68,77,104,113,119,124,133,142],"use":[24,90],"discretized":[26,99,145],"features":[28,40,57,101,147],"has":[29],"gained":[30],"attention":[31],"an":[33,130,149],"efficient":[34],"compatible":[36],"alternative":[37],"to":[38,47,89],"continuous":[39,127],"for":[41,141],"LLMs.":[42],"This":[43],"is":[44],"mainly":[45],"due":[46],"their":[48],"reduced":[49],"storage":[50],"requirements":[51],"better":[53],"alignment":[54],"these":[56],"with":[58,148],"LLM's":[59,105],"input":[60,106,139],"space.":[61],"However,":[62],"typical":[64],"practice":[65],"freezing":[67],"encoder":[70],"during":[71],"training":[72],"poses":[73],"challenges":[74],"in":[75,98,123],"bridging":[76],"modality":[78],"gap":[79],"between":[80],"text.":[83],"To":[84],"address":[85],"this,":[86],"we":[87],"propose":[88],"mixed-scale":[92],"re-tokenization":[93],"layer,":[94],"integrating":[95],"multiple":[96],"granularities":[97],"directly":[102],"within":[103],"module.":[107],"Our":[108],"experimental":[109],"results":[110],"demonstrated":[111],"that":[112],"proposed":[114],"method":[115],"can":[116],"effectively":[117],"enhance":[118],"performance":[120],"ASR":[122],"setting":[125],"learning":[128],"LLM,":[131],"highlighting":[132],"importance":[134],"meticulously":[137],"designed":[138],"module":[140],"integration":[143],"LLM.":[150]},"counts_by_year":[{"year":2025,"cited_by_count":4},{"year":2024,"cited_by_count":2}],"updated_date":"2025-12-26T23:08:49.675405","created_date":"2025-10-10T00:00:00"}