{"id":"https://openalex.org/W4402112241","doi":"https://doi.org/10.21437/interspeech.2024-404","title":"Towards Effective and Efficient Non-autoregressive Decoding Using Block-based Attention Mask","display_name":"Towards Effective and Efficient Non-autoregressive Decoding Using Block-based Attention Mask","publication_year":2024,"publication_date":"2024-09-01","ids":{"openalex":"https://openalex.org/W4402112241","doi":"https://doi.org/10.21437/interspeech.2024-404"},"language":"en","primary_location":{"id":"doi:10.21437/interspeech.2024-404","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2024-404","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2024","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://nrc-publications.canada.ca/eng/view/ft/?id=9dad387d-f86d-4f4b-9d50-063ed037bd68","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5106407750","display_name":"Tianzi Wang","orcid":"https://orcid.org/0009-0005-5823-3039"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tianzi Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5035857878","display_name":"Xurong Xie","orcid":"https://orcid.org/0000-0002-6714-6296"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xurong Xie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101675941","display_name":"Zhaoqing Li","orcid":"https://orcid.org/0000-0001-8649-4934"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhaoqing Li","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5045355404","display_name":"Shoukang Hu","orcid":"https://orcid.org/0000-0002-3345-6923"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shoukang Hu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075023049","display_name":"Zengrui Jin","orcid":"https://orcid.org/0000-0002-2637-7880"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zengrui Jin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5089561793","display_name":"Jiajun Deng","orcid":"https://orcid.org/0000-0001-9624-7451"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiajun Deng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5065896539","display_name":"Mingyu Cui","orcid":"https://orcid.org/0000-0001-9781-9407"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mingyu Cui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5026334377","display_name":"Shujie Hu","orcid":"https://orcid.org/0000-0002-8475-4912"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shujie Hu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003895235","display_name":"Mengzhe Geng","orcid":"https://orcid.org/0000-0002-7886-439X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mengzhe Geng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5042515126","display_name":"Guinan Li","orcid":"https://orcid.org/0000-0002-2206-0237"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Guinan Li","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100969203","display_name":"Helen Meng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Helen Meng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5037109470","display_name":"Xunying Liu","orcid":"https://orcid.org/0000-0001-6725-1160"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xunying Liu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"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":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"262","last_page":"266"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11447","display_name":"Blind Source Separation Techniques","score":0.9873999953269958,"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/T11447","display_name":"Blind Source Separation Techniques","score":0.9873999953269958,"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/T12357","display_name":"Digital Media Forensic Detection","score":0.9789999723434448,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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":0.9656999707221985,"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/autoregressive-model","display_name":"Autoregressive model","score":0.7896968722343445},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.7310308218002319},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6781856417655945},{"id":"https://openalex.org/keywords/block","display_name":"Block (permutation group theory)","score":0.6015601754188538},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3342275619506836},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.32250267267227173},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.1388411521911621},{"id":"https://openalex.org/keywords/econometrics","display_name":"Econometrics","score":0.09129416942596436}],"concepts":[{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.7896968722343445},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.7310308218002319},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6781856417655945},{"id":"https://openalex.org/C2777210771","wikidata":"https://www.wikidata.org/wiki/Q4927124","display_name":"Block (permutation group theory)","level":2,"score":0.6015601754188538},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3342275619506836},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.32250267267227173},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.1388411521911621},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.09129416942596436},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.21437/interspeech.2024-404","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2024-404","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2024","raw_type":"proceedings-article"},{"id":"pmh:oai:cisti-icist.nrc-cnrc.ca:cistinparc:9dad387d-f86d-4f4b-9d50-063ed037bd68","is_oa":true,"landing_page_url":"https://nrc-publications.canada.ca/eng/view/ft/?id=9dad387d-f86d-4f4b-9d50-063ed037bd68","pdf_url":"https://nrc-publications.canada.ca/eng/view/ft/?id=9dad387d-f86d-4f4b-9d50-063ed037bd68","source":{"id":"https://openalex.org/S7407055245","display_name":"NPARC","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"article"}],"best_oa_location":{"id":"pmh:oai:cisti-icist.nrc-cnrc.ca:cistinparc:9dad387d-f86d-4f4b-9d50-063ed037bd68","is_oa":true,"landing_page_url":"https://nrc-publications.canada.ca/eng/view/ft/?id=9dad387d-f86d-4f4b-9d50-063ed037bd68","pdf_url":"https://nrc-publications.canada.ca/eng/view/ft/?id=9dad387d-f86d-4f4b-9d50-063ed037bd68","source":{"id":"https://openalex.org/S7407055245","display_name":"NPARC","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"article"},"sustainable_development_goals":[{"score":0.6299999952316284,"display_name":"Reduced inequalities","id":"https://metadata.un.org/sdg/10"}],"awards":[{"id":"https://openalex.org/G5036096801","display_name":null,"funder_award_id":"2023119","funder_id":"https://openalex.org/F4320322847","funder_display_name":"Youth Innovation Promotion Association of the Chinese Academy of Sciences"},{"id":"https://openalex.org/G6143720773","display_name":null,"funder_award_id":"62106255","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320322847","display_name":"Youth Innovation Promotion Association of the Chinese Academy of Sciences","ror":"https://ror.org/031141b54"},{"id":"https://openalex.org/F4320335892","display_name":"Youth Innovation Promotion Association","ror":null}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4402112241.pdf","grobid_xml":"https://content.openalex.org/works/W4402112241.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2051487156","https://openalex.org/W2073681303","https://openalex.org/W2171218219","https://openalex.org/W1972271943","https://openalex.org/W2150410159","https://openalex.org/W4327525404","https://openalex.org/W4287185323","https://openalex.org/W3150905897","https://openalex.org/W1520183331","https://openalex.org/W2734842993"],"abstract_inverted_index":{"This":[0],"paper":[1],"proposes":[2],"a":[3,57,81],"novel":[4],"non-autoregressive":[5],"(NAR)":[6],"block-based":[7],"Attention":[8],"Mask":[9],"Decoder":[10,75],"(AMD)":[11],"that":[12,32],"flexibly":[13],"balances":[14],"performance-efficiency":[15],"trade-offs":[16],"for":[17],"Conformer":[18],"ASR":[19],"systems.":[20],"AMD":[21,65,78],"performs":[22],"parallel":[23],"NAR":[24],"inference":[25],"within":[26],"contiguous":[27],"blocks":[28],"of":[29,60,86,120],"output":[30],"labels":[31],"are":[33],"concealed":[34],"using":[35],"attention":[36],"masks,":[37],"while":[38,93],"conducting":[39],"left-to-right":[40],"AR":[41,62],"prediction":[42],"and":[43,64,124,128],"history":[44],"context":[45],"amalgamation":[46],"between":[47],"blocks.":[48],"A":[49],"beam":[50],"search":[51],"algorithm":[52],"is":[53],"designed":[54],"to":[55,122],"leverage":[56],"dynamic":[58],"fusion":[59],"CTC,":[61],"Decoder,":[63],"probabilities.":[66],"Experiments":[67],"on":[68,103],"the":[69,73,77,89,104,110,134],"LibriSpeech-100hr":[70],"corpus":[71],"suggest":[72],"tripartite":[74],"incorporating":[76],"module":[79],"produces":[80],"maximum":[82],"decoding":[83,112],"speed-up":[84],"ratio":[85],"1.73x":[87],"over":[88,133],"baseline":[90],"CTC+AR":[91,135],"decoding,":[92],"incurring":[94],"no":[95],"statistically":[96,116],"significant":[97,117],"word":[98],"error":[99],"rate":[100],"(WER)":[101],"increase":[102],"test":[105],"sets.":[106],"When":[107],"operating":[108],"with":[109],"same":[111],"real":[113],"time":[114],"factors,":[115],"WER":[118],"reductions":[119],"up":[121],"0.7%":[123],"0.3%":[125],"absolute":[126],"(5.3%":[127],"6.1%":[129],"relative)":[130],"were":[131],"obtained":[132],"baseline.":[136]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}