{"id":"https://openalex.org/W4375869286","doi":"https://doi.org/10.1109/icassp49357.2023.10096978","title":"Cross-Training: A Semi-Supervised Training Scheme for Speech Recognition","display_name":"Cross-Training: A Semi-Supervised Training Scheme for Speech Recognition","publication_year":2023,"publication_date":"2023-05-05","ids":{"openalex":"https://openalex.org/W4375869286","doi":"https://doi.org/10.1109/icassp49357.2023.10096978"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49357.2023.10096978","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49357.2023.10096978","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-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/A5025206583","display_name":"Soheil Khorram","orcid":"https://orcid.org/0000-0002-2306-3341"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Soheil Khorram","raw_affiliation_strings":["Google Inc.,USA","Google Inc., USA"],"affiliations":[{"raw_affiliation_string":"Google Inc.,USA","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"Google Inc., USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101403998","display_name":"Anshuman Tripathi","orcid":"https://orcid.org/0000-0002-4902-3719"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Anshuman Tripathi","raw_affiliation_strings":["Google Inc.,USA","Google Inc., USA"],"affiliations":[{"raw_affiliation_string":"Google Inc.,USA","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"Google Inc., USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100344655","display_name":"Jaeyoung Kim","orcid":"https://orcid.org/0000-0002-4830-0454"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jaeyoung Kim","raw_affiliation_strings":["Google Inc.,USA","Google Inc., USA"],"affiliations":[{"raw_affiliation_string":"Google Inc.,USA","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"Google Inc., USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5086894713","display_name":"Lu Han","orcid":"https://orcid.org/0000-0002-1120-3220"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Han Lu","raw_affiliation_strings":["Google Inc.,USA","Google Inc., USA"],"affiliations":[{"raw_affiliation_string":"Google Inc.,USA","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"Google Inc., USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114378267","display_name":"Qian Zhang","orcid":"https://orcid.org/0000-0001-9205-1881"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Qian Zhang","raw_affiliation_strings":["Google Inc.,USA","Google Inc., USA"],"affiliations":[{"raw_affiliation_string":"Google Inc.,USA","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"Google Inc., USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032640894","display_name":"Rohit Prabhavalkar","orcid":"https://orcid.org/0000-0001-5331-6058"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Rohit Prabhavalkar","raw_affiliation_strings":["Google Inc.,USA","Google Inc., USA"],"affiliations":[{"raw_affiliation_string":"Google Inc.,USA","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"Google Inc., USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5108537573","display_name":"Ha\u015fim Sak","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hasim Sak","raw_affiliation_strings":["Google Inc.,USA","Google Inc., USA"],"affiliations":[{"raw_affiliation_string":"Google Inc.,USA","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"Google Inc., USA","institution_ids":["https://openalex.org/I1291425158"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5025206583"],"corresponding_institution_ids":["https://openalex.org/I1291425158"],"apc_list":null,"apc_paid":null,"fwci":0.5237,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.69778676,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":96},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"5"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9998999834060669,"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":0.9998999834060669,"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":0.9976000189781189,"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":0.9947999715805054,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.8384732007980347},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7551705837249756},{"id":"https://openalex.org/keywords/scheme","display_name":"Scheme (mathematics)","score":0.6802242398262024},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.6404263377189636},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.44172075390815735},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.43753620982170105},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.3203628659248352},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.057302117347717285}],"concepts":[{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.8384732007980347},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7551705837249756},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.6802242398262024},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.6404263377189636},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.44172075390815735},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.43753620982170105},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3203628659248352},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.057302117347717285},{"id":"https://openalex.org/C153294291","wikidata":"https://www.wikidata.org/wiki/Q25261","display_name":"Meteorology","level":1,"score":0.0},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49357.2023.10096978","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49357.2023.10096978","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":27,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W1828163288","https://openalex.org/W2151742940","https://openalex.org/W2936774411","https://openalex.org/W2964110616","https://openalex.org/W2972943112","https://openalex.org/W2973049979","https://openalex.org/W3008181812","https://openalex.org/W3016010032","https://openalex.org/W3016234571","https://openalex.org/W3016970897","https://openalex.org/W3036601975","https://openalex.org/W3041561163","https://openalex.org/W3097777922","https://openalex.org/W3103934428","https://openalex.org/W3160799772","https://openalex.org/W3163203022","https://openalex.org/W3204696009","https://openalex.org/W3209984917","https://openalex.org/W3212799896","https://openalex.org/W4221145109","https://openalex.org/W4221161761","https://openalex.org/W4224919704","https://openalex.org/W4225307083","https://openalex.org/W4225755266","https://openalex.org/W4226033575","https://openalex.org/W4297808394"],"related_works":["https://openalex.org/W230091440","https://openalex.org/W2233261550","https://openalex.org/W2810751659","https://openalex.org/W258997015","https://openalex.org/W2997094352","https://openalex.org/W3216976533","https://openalex.org/W100620283","https://openalex.org/W2047028599","https://openalex.org/W4394050964","https://openalex.org/W2551249631"],"abstract_inverted_index":{"Semi-supervised":[0],"training":[1,36,137],"can":[2,89],"be":[3,85],"performed":[4],"by":[5,60],"jointly":[6],"optimizing":[7],"supervised":[8,15,145],"and":[9,16,21,112,138],"unsupervised":[10,17],"losses.":[11],"In":[12],"many":[13],"settings,":[14],"losses":[18],"are":[19],"inconsistent,":[20],"this":[22],"inconsistency":[23],"creates":[24],"instability":[25],"in":[26],"training.":[27],"As":[28],"a":[29,49,74,102],"solution,":[30],"we":[31,42,52],"propose":[32],"cross-training:":[33],"instead":[34],"of":[35,57],"one":[37],"network":[38],"with":[39,48],"two":[40,44],"losses,":[41],"train":[43],"separate":[45],"networks,":[46],"each":[47,92],"different":[50],"loss;":[51],"then":[53],"tie":[54],"the":[55,58,67,78,82,97,126,130,155],"parameters":[56],"networks":[59,83],"minimizing":[61],"an":[62],"additional":[63],"L2":[64,70],"loss":[65,71],"between":[66,77],"parameters.":[68],"This":[69,94],"acts":[72],"as":[73],"knowledge":[75],"bridge":[76],"networks.":[79],"It":[80],"forces":[81],"to":[84,100],"similar;":[86],"therefore":[87],"both":[88],"learn":[90],"from":[91],"other.":[93],"paper":[95],"introduces":[96],"cross-training":[98,119,134,148],"scheme":[99],"develop":[101],"stable":[103],"contrastive":[104],"siamese":[105],"(c-siam)":[106],"network.":[107],"Our":[108],"experiments":[109],"on":[110,129,143],"LibriSpeech":[111,131],"Google\u2019s":[113],"Voice-Search/YouTube":[114],"datasets":[115],"show":[116],"that":[117],"(1)":[118],"provides":[120],"20%":[121],"relative":[122],"WER":[123],"improvement":[124],"over":[125],"SOTA":[127,141],"systems":[128,142],"dataset;":[132],"(2)":[133],"stabilizes":[135],"c-siam":[136,157],"significantly":[139],"outperforms":[140],"small":[144],"datasets;":[146],"(3)":[147],"is":[149],"effective":[150],"for":[151],"cascaded":[152],"encoders,":[153],"unlike":[154],"original":[156],"which":[158],"shows":[159],"weak":[160],"convergence":[161],"characteristics.":[162]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2023,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}