{"id":"https://openalex.org/W4414955889","doi":"https://doi.org/10.1109/tnnls.2025.3615971","title":"IML-Spikeformer: Input-Aware Multilevel Spiking Transformer for Speech Processing","display_name":"IML-Spikeformer: Input-Aware Multilevel Spiking Transformer for Speech Processing","publication_year":2025,"publication_date":"2025-10-08","ids":{"openalex":"https://openalex.org/W4414955889","doi":"https://doi.org/10.1109/tnnls.2025.3615971","pmid":"https://pubmed.ncbi.nlm.nih.gov/41060856"},"language":"en","primary_location":{"id":"doi:10.1109/tnnls.2025.3615971","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tnnls.2025.3615971","pdf_url":null,"source":{"id":"https://openalex.org/S4210175523","display_name":"IEEE Transactions on Neural Networks and Learning Systems","issn_l":"2162-237X","issn":["2162-237X","2162-2388"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Neural Networks and Learning Systems","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","pubmed"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Zeyang Song","orcid":"https://orcid.org/0009-0006-4162-3046"},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Zeyang Song","raw_affiliation_strings":["Department of Electrical and Computer Engineering, National University of Singapore, Queenstown, Singapore"],"raw_orcid":"https://orcid.org/0009-0006-4162-3046","affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, National University of Singapore, Queenstown, Singapore","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100630316","display_name":"Shimin Zhang","orcid":"https://orcid.org/0009-0004-4344-5218"},"institutions":[{"id":"https://openalex.org/I14243506","display_name":"Hong Kong Polytechnic University","ror":"https://ror.org/0030zas98","country_code":"HK","type":"education","lineage":["https://openalex.org/I14243506"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Shimin Zhang","raw_affiliation_strings":["Department of Data Science and Artificial Intelligence, The Hong Kong Polytechnic University, Hung Hom, SAR, Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Data Science and Artificial Intelligence, The Hong Kong Polytechnic University, Hung Hom, SAR, Hong Kong","institution_ids":["https://openalex.org/I14243506"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5113151651","display_name":"Yuhong Chou","orcid":null},"institutions":[{"id":"https://openalex.org/I14243506","display_name":"Hong Kong Polytechnic University","ror":"https://ror.org/0030zas98","country_code":"HK","type":"education","lineage":["https://openalex.org/I14243506"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Yuhong Chou","raw_affiliation_strings":["Department of Data Science and Artificial Intelligence, The Hong Kong Polytechnic University, Hung Hom, SAR, Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Data Science and Artificial Intelligence, The Hong Kong Polytechnic University, Hung Hom, SAR, Hong Kong","institution_ids":["https://openalex.org/I14243506"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018341980","display_name":"Jibin Wu","orcid":"https://orcid.org/0000-0003-0135-4188"},"institutions":[{"id":"https://openalex.org/I14243506","display_name":"Hong Kong Polytechnic University","ror":"https://ror.org/0030zas98","country_code":"HK","type":"education","lineage":["https://openalex.org/I14243506"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Jibin Wu","raw_affiliation_strings":["Department of Data Science and Artificial Intelligence, Department of Computing, The Hong Kong Polytechnic University, Hung Hom, SAR, Hong Kong"],"raw_orcid":"https://orcid.org/0000-0003-0135-4188","affiliations":[{"raw_affiliation_string":"Department of Data Science and Artificial Intelligence, Department of Computing, The Hong Kong Polytechnic University, Hung Hom, SAR, Hong Kong","institution_ids":["https://openalex.org/I14243506"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5032690182","display_name":"Haizhou Li","orcid":"https://orcid.org/0000-0001-9158-9401"},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Haizhou Li","raw_affiliation_strings":["Department of Electrical and Computer Engineering, National University of Singapore, Queenstown, Singapore"],"raw_orcid":"https://orcid.org/0000-0001-9158-9401","affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, National University of Singapore, Queenstown, Singapore","institution_ids":["https://openalex.org/I165932596"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":2.7073,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.91251959,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":97},"biblio":{"volume":"37","issue":"3","first_page":"1377","last_page":"1389"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":0.9957000017166138,"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":0.9957000017166138,"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":0.9674000144004822,"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/neuromorphic-engineering","display_name":"Neuromorphic engineering","score":0.7480000257492065},{"id":"https://openalex.org/keywords/spiking-neural-network","display_name":"Spiking neural network","score":0.6488000154495239},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.5906999707221985},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5482000112533569},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.5465999841690063},{"id":"https://openalex.org/keywords/spike","display_name":"Spike (software development)","score":0.5464000105857849},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5321999788284302},{"id":"https://openalex.org/keywords/speech-processing","display_name":"Speech processing","score":0.48019999265670776}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8077999949455261},{"id":"https://openalex.org/C151927369","wikidata":"https://www.wikidata.org/wiki/Q1981312","display_name":"Neuromorphic engineering","level":3,"score":0.7480000257492065},{"id":"https://openalex.org/C11731999","wikidata":"https://www.wikidata.org/wiki/Q9067355","display_name":"Spiking neural network","level":3,"score":0.6488000154495239},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5906999707221985},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5482000112533569},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.5465999841690063},{"id":"https://openalex.org/C2781390188","wikidata":"https://www.wikidata.org/wiki/Q25203449","display_name":"Spike (software development)","level":2,"score":0.5464000105857849},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5321999788284302},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.48019999265670776},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.4456000030040741},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.421999990940094},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.41040000319480896},{"id":"https://openalex.org/C191178318","wikidata":"https://www.wikidata.org/wiki/Q2256906","display_name":"Thresholding","level":3,"score":0.323199987411499},{"id":"https://openalex.org/C2776182073","wikidata":"https://www.wikidata.org/wiki/Q7575395","display_name":"Speech enhancement","level":3,"score":0.30869999527931213},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.30799999833106995},{"id":"https://openalex.org/C2780165032","wikidata":"https://www.wikidata.org/wiki/Q16869822","display_name":"Energy consumption","level":2,"score":0.30219998955726624},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.2955999970436096},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.2928999960422516},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.2615000009536743},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.25769999623298645},{"id":"https://openalex.org/C13895895","wikidata":"https://www.wikidata.org/wiki/Q3270773","display_name":"Speech coding","level":2,"score":0.2563999891281128},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.25130000710487366}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.1109/tnnls.2025.3615971","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tnnls.2025.3615971","pdf_url":null,"source":{"id":"https://openalex.org/S4210175523","display_name":"IEEE Transactions on Neural Networks and Learning Systems","issn_l":"2162-237X","issn":["2162-237X","2162-2388"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Neural Networks and Learning Systems","raw_type":"journal-article"},{"id":"pmid:41060856","is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/41060856","pdf_url":null,"source":{"id":"https://openalex.org/S4306525036","display_name":"PubMed","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1299303238","host_organization_name":"National Institutes of Health","host_organization_lineage":["https://openalex.org/I1299303238"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE transactions on neural networks and learning systems","raw_type":null},{"id":"pmh:oai:ira.lib.polyu.edu.hk:10397/117186","is_oa":false,"landing_page_url":"http://hdl.handle.net/10397/117186","pdf_url":null,"source":{"id":"https://openalex.org/S4306400205","display_name":"PolyU Institutional Research Archive (Hong Kong Polytechnic University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I14243506","host_organization_name":"Hong Kong Polytechnic University","host_organization_lineage":["https://openalex.org/I14243506"],"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":"Journal/Magazine Article"},{"id":"pmh:oai:ira.lib.polyu.edu.hk:10397/117407","is_oa":false,"landing_page_url":"http://hdl.handle.net/10397/117407","pdf_url":null,"source":{"id":"https://openalex.org/S4306400205","display_name":"PolyU Institutional Research Archive (Hong Kong Polytechnic University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I14243506","host_organization_name":"Hong Kong Polytechnic University","host_organization_lineage":["https://openalex.org/I14243506"],"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":"Journal/Magazine Article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G2159245610","display_name":null,"funder_award_id":"62306259","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G2799040894","display_name":null,"funder_award_id":"62271432","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"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Spiking":[0],"neural":[1,7,22],"networks":[2,23],"(SNNs),":[3],"inspired":[4],"by":[5,54,183],"biological":[6],"mechanisms,":[8],"represent":[9],"a":[10,82,109,121,128],"promising":[11],"neuromorphic":[12],"computing":[13],"paradigm":[14],"that":[15,157],"offers":[16],"energy-efficient":[17],"alternatives":[18],"to":[19,32,67,93,173],"traditional":[20],"artificial":[21],"(ANNs).":[24],"Despite":[25],"proven":[26],"effectiveness,":[27],"SNN":[28,64,197],"architectures":[29,65,198],"have":[30],"struggled":[31],"achieve":[33],"competitive":[34],"performance":[35,206],"on":[36,166,170],"large-scale":[37,63,89,200],"speech":[38,68,90,153,201],"processing":[39,69,202],"tasks.":[40,70],"Two":[41],"key":[42],"challenges":[43],"hinder":[44],"progress:":[45],"1)":[46],"the":[47,60,73,77,97,134,140],"high":[48],"computational":[49],"overhead":[50],"during":[51],"training":[52],"caused":[53],"multitimestep":[55,105],"spike":[56,100,106],"firing":[57,107],"and":[58,145,168,186,207,213],"2)":[59],"absence":[61],"of":[62,142,148,164,195],"tailored":[66],"To":[71],"overcome":[72],"issues,":[74],"we":[75],"introduce":[76],"input-aware":[78,98,115],"multilevel":[79,99],"spikeformer":[80],"(IML-Spikeformer),":[81],"spiking":[83,123],"transformer":[84],"architecture":[85],"specifically":[86],"designed":[87],"for":[88,199],"processing.":[91],"Central":[92],"our":[94],"design":[95],"is":[96],"(IMLS)":[101],"mechanism,":[102],"which":[103],"simulates":[104],"within":[108],"single":[110],"timestep":[111],"using":[112],"an":[113,193],"adaptive,":[114],"thresholding":[116],"scheme.":[117],"IML-Spikeformer":[118,158,191],"further":[119],"integrates":[120],"reparameterized":[122],"self-attention":[124],"(RepSSA)":[125],"module":[126,138],"with":[127],"hierarchical":[129],"decay":[130],"mask":[131],"(HDM),":[132],"forming":[133],"HD-RepSSA":[135],"module.":[136],"This":[137],"enhances":[139],"precision":[141],"attention":[143],"maps":[144],"enables":[146],"modeling":[147],"multiscale":[149],"temporal":[150],"dependencies":[151],"in":[152,203],"signals.":[154],"Experiments":[155],"demonstrate":[156],"achieves":[159],"word":[160],"error":[161],"rates":[162],"(WERs)":[163],"6.0%":[165],"AiShell-1":[167],"3.4%":[169],"Librispeech-960,":[171],"comparable":[172],"conventional":[174],"ANN":[175],"transformers":[176],"while":[177],"reducing":[178],"theoretical":[179],"inference":[180],"energy":[181,208],"consumption":[182],"$4.64\\times":[184],"$":[185,188],"$4.32\\times":[187],",":[189],"respectively.":[190],"marks":[192],"advance":[194],"scalable":[196],"both":[204],"task":[205],"efficiency.":[209],"Our":[210],"source":[211],"code":[212],"model":[214],"checkpoints":[215],"are":[216],"publicly":[217],"available":[218],"at":[219],"github.com/Pooookeman/IML-Spikeformer.":[220]},"counts_by_year":[{"year":2025,"cited_by_count":3}],"updated_date":"2026-03-03T06:13:14.889584","created_date":"2025-10-10T00:00:00"}