{"id":"https://openalex.org/W4392909513","doi":"https://doi.org/10.1109/icassp48485.2024.10447447","title":"Improving Kinyarwanda Speech Recognition Via Semi-Supervised Learning","display_name":"Improving Kinyarwanda Speech Recognition Via Semi-Supervised Learning","publication_year":2024,"publication_date":"2024-03-18","ids":{"openalex":"https://openalex.org/W4392909513","doi":"https://doi.org/10.1109/icassp48485.2024.10447447"},"language":"en","primary_location":{"id":"doi:10.1109/icassp48485.2024.10447447","is_oa":true,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10447447","pdf_url":"https://ieeexplore.ieee.org/ielx7/10445798/10445803/10447447.pdf","source":null,"license":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://ieeexplore.ieee.org/ielx7/10445798/10445803/10447447.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5087847514","display_name":"Antoine Nzeyimana","orcid":null},"institutions":[{"id":"https://openalex.org/I24603500","display_name":"University of Massachusetts Amherst","ror":"https://ror.org/0072zz521","country_code":"US","type":"education","lineage":["https://openalex.org/I24603500"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Antoine Nzeyimana","raw_affiliation_strings":["University of Massachusetts,Amherst,USA","University of Massachusetts, Amherst, USA"],"affiliations":[{"raw_affiliation_string":"University of Massachusetts,Amherst,USA","institution_ids":["https://openalex.org/I24603500"]},{"raw_affiliation_string":"University of Massachusetts, Amherst, USA","institution_ids":["https://openalex.org/I24603500"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5087847514"],"corresponding_institution_ids":["https://openalex.org/I24603500"],"apc_list":null,"apc_paid":null,"fwci":0.7501,"has_fulltext":true,"cited_by_count":2,"citation_normalized_percentile":{"value":0.65467395,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"10701","last_page":"10705"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":0.9402999877929688,"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"}},"topics":[{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":0.9402999877929688,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.9218999743461609,"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":0.7741333246231079},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.7018833160400391},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4521227180957794},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.364992618560791}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7741333246231079},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.7018833160400391},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4521227180957794},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.364992618560791}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp48485.2024.10447447","is_oa":true,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10447447","pdf_url":"https://ieeexplore.ieee.org/ielx7/10445798/10445803/10447447.pdf","source":null,"license":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1109/icassp48485.2024.10447447","is_oa":true,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10447447","pdf_url":"https://ieeexplore.ieee.org/ielx7/10445798/10445803/10447447.pdf","source":null,"license":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4392909513.pdf","grobid_xml":"https://content.openalex.org/works/W4392909513.grobid-xml"},"referenced_works_count":24,"referenced_works":["https://openalex.org/W126677248","https://openalex.org/W2127141656","https://openalex.org/W2296073425","https://openalex.org/W2963390466","https://openalex.org/W2963537349","https://openalex.org/W2974231335","https://openalex.org/W2978923831","https://openalex.org/W3036601975","https://openalex.org/W3037469336","https://openalex.org/W3093579165","https://openalex.org/W3097777922","https://openalex.org/W3167533889","https://openalex.org/W4210984894","https://openalex.org/W4295312788","https://openalex.org/W4300416243","https://openalex.org/W6605205965","https://openalex.org/W6766978945","https://openalex.org/W6767671539","https://openalex.org/W6768952226","https://openalex.org/W6771467084","https://openalex.org/W6779902714","https://openalex.org/W6780218876","https://openalex.org/W6784614252","https://openalex.org/W6844712592"],"related_works":["https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W2358668433","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W2382290278","https://openalex.org/W2478288626","https://openalex.org/W4391913857","https://openalex.org/W2350741829","https://openalex.org/W3204019825"],"abstract_inverted_index":{"Achieving":[0],"robust":[1],"speech":[2,31,47,128],"recognition":[3,32,129],"for":[4,33,68,130],"Kinyarwanda":[5],"remains":[6],"a":[7,21,52,63,71],"challenging":[8],"task.":[9],"In":[10],"this":[11],"work,":[12],"we":[13],"empirically":[14],"show":[15],"that":[16,108,118],"using":[17,27,39,119],"self-supervised":[18,112],"pretraining,":[19],"following":[20],"curriculum":[22,65],"schedule":[23,67],"during":[24],"fine-tuning":[25],"and":[26,58,94],"semi-supervised":[28,79],"learning":[29,66],"improve":[30,104],"Kinyarwanda.":[34,131],"Our":[35,114],"approach":[36],"focuses":[37],"on":[38,70,90,97],"public":[40,53],"domain":[41],"data":[42],"only.":[43],"A":[44],"new":[45,92],"studio-quality":[46],"dataset":[48,93],"is":[49],"collected":[50],"from":[51],"website,":[54],"aligned":[55],"to":[56,61],"text":[57],"then":[59],"used":[60],"formulate":[62],"simple":[64],"training":[69],"larger,":[72],"noisier":[73],"dataset.":[74],"After":[75],"four":[76],"generations":[77],"of":[78],"learning,":[80],"our":[81],"final":[82],"model":[83],"achieves":[84],"3.2%":[85],"word":[86],"error":[87],"rate":[88],"(WER)":[89],"the":[91],"15.9%":[95],"WER":[96],"Mozilla":[98],"Common":[99],"Voice":[100],"benchmark.":[101],"These":[102],"results":[103,125],"upon":[105],"off-the-shelf":[106],"models":[107],"use":[109],"English":[110],"language":[111],"representations.":[113],"experiments":[115],"also":[116],"indicate":[117],"syllabic":[120],"rather":[121],"than":[122],"character-based":[123],"tokenization":[124],"in":[126],"better":[127]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1}],"updated_date":"2025-11-25T21:42:39.735039","created_date":"2025-10-10T00:00:00"}