{"id":"https://openalex.org/W4405709181","doi":"https://doi.org/10.1109/iscslp63861.2024.10800511","title":"ConMamba: A Convolution-Augmented Mamba Encoder Model for Efficient End-to-End ASR Systems","display_name":"ConMamba: A Convolution-Augmented Mamba Encoder Model for Efficient End-to-End ASR Systems","publication_year":2024,"publication_date":"2024-11-07","ids":{"openalex":"https://openalex.org/W4405709181","doi":"https://doi.org/10.1109/iscslp63861.2024.10800511"},"language":"en","primary_location":{"id":"doi:10.1109/iscslp63861.2024.10800511","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iscslp63861.2024.10800511","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":"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/A5108960067","display_name":"Haoxiang Hou","orcid":null},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Haoxiang Hou","raw_affiliation_strings":["AI Institute, Shanghai Jiao Tong University,Auditory Cognition and Computational Acoustics Lab MoE, Key Lab of Artificial Intelligence,Department of Computer Science and Engineering,Shanghai"],"affiliations":[{"raw_affiliation_string":"AI Institute, Shanghai Jiao Tong University,Auditory Cognition and Computational Acoustics Lab MoE, Key Lab of Artificial Intelligence,Department of Computer Science and Engineering,Shanghai","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5086338661","display_name":"Xun Gong","orcid":"https://orcid.org/0000-0002-3364-8407"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xun Gong","raw_affiliation_strings":["AI Institute, Shanghai Jiao Tong University,Auditory Cognition and Computational Acoustics Lab MoE, Key Lab of Artificial Intelligence,Department of Computer Science and Engineering,Shanghai"],"affiliations":[{"raw_affiliation_string":"AI Institute, Shanghai Jiao Tong University,Auditory Cognition and Computational Acoustics Lab MoE, Key Lab of Artificial Intelligence,Department of Computer Science and Engineering,Shanghai","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100341993","display_name":"Yanmin Qian","orcid":"https://orcid.org/0000-0002-0314-3790"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yanmin Qian","raw_affiliation_strings":["AI Institute, Shanghai Jiao Tong University,Auditory Cognition and Computational Acoustics Lab MoE, Key Lab of Artificial Intelligence,Department of Computer Science and Engineering,Shanghai"],"affiliations":[{"raw_affiliation_string":"AI Institute, Shanghai Jiao Tong University,Auditory Cognition and Computational Acoustics Lab MoE, Key Lab of Artificial Intelligence,Department of Computer Science and Engineering,Shanghai","institution_ids":["https://openalex.org/I183067930"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5108960067"],"corresponding_institution_ids":["https://openalex.org/I183067930"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.27079987,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"711","last_page":"715"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":0.9958999752998352,"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":0.9958999752998352,"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/T11447","display_name":"Blind Source Separation Techniques","score":0.9818000197410583,"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":0.9645000100135803,"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/encoder","display_name":"Encoder","score":0.7381669282913208},{"id":"https://openalex.org/keywords/convolution","display_name":"Convolution (computer science)","score":0.7006098031997681},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6517794132232666},{"id":"https://openalex.org/keywords/end-to-end-principle","display_name":"End-to-end principle","score":0.4123958945274353},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.1452321708202362},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.10252973437309265}],"concepts":[{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.7381669282913208},{"id":"https://openalex.org/C45347329","wikidata":"https://www.wikidata.org/wiki/Q5166604","display_name":"Convolution (computer science)","level":3,"score":0.7006098031997681},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6517794132232666},{"id":"https://openalex.org/C74296488","wikidata":"https://www.wikidata.org/wiki/Q2527392","display_name":"End-to-end principle","level":2,"score":0.4123958945274353},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.1452321708202362},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.10252973437309265},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/iscslp63861.2024.10800511","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iscslp63861.2024.10800511","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":[],"awards":[{"id":"https://openalex.org/G5465080215","display_name":null,"funder_award_id":"2021SHZDZX0102","funder_id":"https://openalex.org/F4320321885","funder_display_name":"Science and Technology Commission of Shanghai Municipality"}],"funders":[{"id":"https://openalex.org/F4320321885","display_name":"Science and Technology Commission of Shanghai Municipality","ror":"https://ror.org/03kt66j61"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":20,"referenced_works":["https://openalex.org/W2327501763","https://openalex.org/W2892009249","https://openalex.org/W2972818416","https://openalex.org/W3016010032","https://openalex.org/W3097777922","https://openalex.org/W4399939359","https://openalex.org/W6629717138","https://openalex.org/W6638749077","https://openalex.org/W6739901393","https://openalex.org/W6802912272","https://openalex.org/W6803444062","https://openalex.org/W6809915218","https://openalex.org/W6839026989","https://openalex.org/W6839235732","https://openalex.org/W6859298233","https://openalex.org/W6860720708","https://openalex.org/W6861387779","https://openalex.org/W6862687296","https://openalex.org/W6863493712","https://openalex.org/W6863679278"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2151749779","https://openalex.org/W3179968364","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109"],"abstract_inverted_index":{"End-to-End":[0],"Automatic":[1],"Speech":[2],"Recognition":[3],"(ASR)":[4],"models,":[5],"such":[6],"as":[7],"Conformer,":[8],"excel":[9],"in":[10,15,24],"accuracy":[11],"but":[12],"face":[13],"limitations":[14],"computational":[16],"complexity":[17],"and":[18,63,71,89,97],"positional":[19],"awareness,":[20],"hindering":[21],"their":[22],"use":[23],"real-time":[25],"or":[26],"resource-constrained":[27],"settings.":[28],"State":[29],"Space":[30],"Models":[31],"(SSMs),":[32],"particularly":[33],"the":[34,56,76],"Mamba":[35,51,61],"model":[36],"with":[37,60],"its":[38],"time-varying":[39],"mechanism,":[40],"offer":[41],"a":[42,49],"more":[43],"efficient":[44],"alternative.":[45],"We":[46],"propose":[47],"ConMamba,":[48],"Convolution-Augmented":[50],"En-coder":[52],"model,":[53],"which":[54],"replaces":[55],"Conformer's":[57],"Multi-Head":[58],"Self-Attention":[59],"layers":[62,66],"adds":[64],"convolutional":[65],"for":[67,99],"capturing":[68],"both":[69],"local":[70],"global":[72],"features.":[73],"Experiments":[74],"on":[75,85,92],"LibriSpeech":[77],"dataset":[78],"show":[79],"ConMamba":[80],"matches":[81],"traditional":[82],"Conformer":[83],"performance":[84],"short":[86],"speech":[87],"segments":[88],"outperforms":[90],"them":[91],"longer":[93],"ones,":[94],"enhancing":[95],"robustness":[96],"efficiency":[98],"practical":[100],"ASR":[101],"applications.":[102]},"counts_by_year":[],"updated_date":"2025-12-27T23:08:20.325037","created_date":"2025-10-10T00:00:00"}