{"id":"https://openalex.org/W4405709454","doi":"https://doi.org/10.1109/iscslp63861.2024.10800477","title":"Bridging Speech and Text: Enhancing ASR with Pinyin-to-Character Pre-training in LLMs","display_name":"Bridging Speech and Text: Enhancing ASR with Pinyin-to-Character Pre-training in LLMs","publication_year":2024,"publication_date":"2024-11-07","ids":{"openalex":"https://openalex.org/W4405709454","doi":"https://doi.org/10.1109/iscslp63861.2024.10800477"},"language":"en","primary_location":{"id":"doi:10.1109/iscslp63861.2024.10800477","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iscslp63861.2024.10800477","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE 14th International Symposium on Chinese Spoken Language Processing (ISCSLP)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5101019493","display_name":"Yang Yuhang","orcid":"https://orcid.org/0009-0006-4413-0482"},"institutions":[{"id":"https://openalex.org/I16609230","display_name":"Hunan University","ror":"https://ror.org/05htk5m33","country_code":"CN","type":"education","lineage":["https://openalex.org/I16609230"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuhang Yang","raw_affiliation_strings":["Hunan University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hunan University","institution_ids":["https://openalex.org/I16609230"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114025990","display_name":"Yizhou Peng","orcid":"https://orcid.org/0000-0002-5718-570X"},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Yizhou Peng","raw_affiliation_strings":["Nanyang Technological University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nanyang Technological University","institution_ids":["https://openalex.org/I172675005"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5070872826","display_name":"Eng Siong Chng","orcid":"https://orcid.org/0000-0001-6257-7399"},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Eng Siong Chng","raw_affiliation_strings":["Nanyang Technological University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nanyang Technological University","institution_ids":["https://openalex.org/I172675005"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100681425","display_name":"Xionghu Zhong","orcid":"https://orcid.org/0000-0002-7533-2347"},"institutions":[{"id":"https://openalex.org/I16609230","display_name":"Hunan University","ror":"https://ror.org/05htk5m33","country_code":"CN","type":"education","lineage":["https://openalex.org/I16609230"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xionghu Zhong","raw_affiliation_strings":["Hunan University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hunan University","institution_ids":["https://openalex.org/I16609230"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"646","last_page":"650"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9944999814033508,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9944999814033508,"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/T12353","display_name":"Lexicography and Language Studies","score":0.9649999737739563,"subfield":{"id":"https://openalex.org/subfields/1203","display_name":"Language and Linguistics"},"field":{"id":"https://openalex.org/fields/12","display_name":"Arts and Humanities"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T11587","display_name":"Second Language Acquisition and Learning","score":0.9363999962806702,"subfield":{"id":"https://openalex.org/subfields/3204","display_name":"Developmental and Educational Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/pinyin","display_name":"Pinyin","score":0.9155595302581787},{"id":"https://openalex.org/keywords/bridging","display_name":"Bridging (networking)","score":0.8526110053062439},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6887706518173218},{"id":"https://openalex.org/keywords/character","display_name":"Character (mathematics)","score":0.6800522804260254},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.48188358545303345},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.42375749349594116},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.306643009185791},{"id":"https://openalex.org/keywords/chinese-characters","display_name":"Chinese characters","score":0.13663864135742188},{"id":"https://openalex.org/keywords/computer-security","display_name":"Computer security","score":0.05590376257896423},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.05144563317298889}],"concepts":[{"id":"https://openalex.org/C2781095461","wikidata":"https://www.wikidata.org/wiki/Q42222","display_name":"Pinyin","level":3,"score":0.9155595302581787},{"id":"https://openalex.org/C174348530","wikidata":"https://www.wikidata.org/wiki/Q188635","display_name":"Bridging (networking)","level":2,"score":0.8526110053062439},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6887706518173218},{"id":"https://openalex.org/C2780861071","wikidata":"https://www.wikidata.org/wiki/Q1062934","display_name":"Character (mathematics)","level":2,"score":0.6800522804260254},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.48188358545303345},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.42375749349594116},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.306643009185791},{"id":"https://openalex.org/C2781051154","wikidata":"https://www.wikidata.org/wiki/Q8201","display_name":"Chinese characters","level":2,"score":0.13663864135742188},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.05590376257896423},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.05144563317298889},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/iscslp63861.2024.10800477","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iscslp63861.2024.10800477","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE 14th International Symposium on Chinese Spoken Language Processing (ISCSLP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Quality Education","score":0.550000011920929,"id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":29,"referenced_works":["https://openalex.org/W2963242190","https://openalex.org/W3030163527","https://openalex.org/W3203407300","https://openalex.org/W3209059054","https://openalex.org/W3209984917","https://openalex.org/W4382246105","https://openalex.org/W4389520395","https://openalex.org/W4389524500","https://openalex.org/W4391021666","https://openalex.org/W4392903956","https://openalex.org/W4392931626","https://openalex.org/W4401023891","https://openalex.org/W6780218876","https://openalex.org/W6796581206","https://openalex.org/W6810081322","https://openalex.org/W6810334672","https://openalex.org/W6847363464","https://openalex.org/W6849177959","https://openalex.org/W6850625674","https://openalex.org/W6851592950","https://openalex.org/W6852447913","https://openalex.org/W6852489829","https://openalex.org/W6853094705","https://openalex.org/W6856794988","https://openalex.org/W6856800273","https://openalex.org/W6857054612","https://openalex.org/W6857589352","https://openalex.org/W6859099255","https://openalex.org/W6861626678"],"related_works":["https://openalex.org/W2355589192","https://openalex.org/W2383579380","https://openalex.org/W2599785913","https://openalex.org/W2357614476","https://openalex.org/W2391413009","https://openalex.org/W2977960882","https://openalex.org/W3004865318","https://openalex.org/W2180791503","https://openalex.org/W3016183518","https://openalex.org/W1712848793"],"abstract_inverted_index":{"The":[0],"integration":[1],"of":[2,98],"large":[3],"language":[4],"models":[5,10],"(LLMs)":[6],"with":[7],"pretrained":[8],"speech":[9,18,85,99],"has":[11],"opened":[12],"up":[13],"new":[14],"avenues":[15],"in":[16,24,49,112],"automatic":[17],"recognition":[19],"(ASR).":[20],"While":[21],"LLMs":[22,55],"excel":[23],"multimodal":[25],"understanding":[26,97],"tasks,":[27],"effectively":[28],"leveraging":[29],"their":[30],"capabilities":[31],"for":[32,127],"ASR":[33,50,113],"remains":[34],"a":[35,41,108,134],"significant":[36],"challenge.":[37],"This":[38,69],"paper":[39],"presents":[40],"novel":[42],"training":[43],"approach":[44,106],"to":[45,64,74,76,93,116],"enhance":[46,94],"LLM":[47,73],"performance":[48],"tasks.":[51],"We":[52],"propose":[53],"pre-training":[54,129],"on":[56],"Pinyin":[57],"embedding":[58],"sequences,":[59],"which":[60],"represent":[61],"pronunciation":[62,80],"features,":[63],"generate":[65],"corresponding":[66],"Chinese":[67],"characters.":[68],"step":[70],"enables":[71],"the":[72,90,95,117],"adapt":[75],"generating":[77],"text":[78,125],"from":[79],"features":[81],"before":[82],"encountering":[83],"real":[84],"data.":[86],"Furthermore,":[87],"we":[88],"fine-tune":[89],"LoRA":[91],"parameters":[92],"LLM's":[96],"modality":[100],"information.":[101],"In":[102],"AISHELL-1":[103],"corpus,":[104],"our":[105],"yields":[107],"9.5%":[109],"relative":[110,136],"improvement":[111],"tasks":[114],"compared":[115],"baseline":[118],"without":[119],"Pinyi-to-Character":[120,128],"pre-training.":[121],"Additionally,":[122],"incorporating":[123],"auxiliary":[124],"data":[126],"further":[130],"boosts":[131],"performance,":[132],"achieving":[133],"19.0%":[135],"improvement.":[137]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}