{"id":"https://openalex.org/W4408352558","doi":"https://doi.org/10.1109/icassp49660.2025.10889560","title":"Attention Weighting and Conditional Entropy-driven Quantization Loss for Neural Audio Codecs","display_name":"Attention Weighting and Conditional Entropy-driven Quantization Loss for Neural Audio Codecs","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408352558","doi":"https://doi.org/10.1109/icassp49660.2025.10889560"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10889560","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10889560","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2025 - 2025 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/A5029591050","display_name":"Huiyu Zhang","orcid":"https://orcid.org/0000-0002-8958-3155"},"institutions":[{"id":"https://openalex.org/I37461747","display_name":"Wuhan University","ror":"https://ror.org/033vjfk17","country_code":"CN","type":"education","lineage":["https://openalex.org/I37461747"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Huiyu Zhang","raw_affiliation_strings":["Wuhan University,NERCMS, School of Computer Science, Hubei Luojia Laboratory,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Wuhan University,NERCMS, School of Computer Science, Hubei Luojia Laboratory,China","institution_ids":["https://openalex.org/I37461747"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041664453","display_name":"Weiping Tu","orcid":"https://orcid.org/0000-0002-6933-3298"},"institutions":[{"id":"https://openalex.org/I37461747","display_name":"Wuhan University","ror":"https://ror.org/033vjfk17","country_code":"CN","type":"education","lineage":["https://openalex.org/I37461747"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Weiping Tu","raw_affiliation_strings":["Wuhan University,NERCMS, School of Computer Science, Hubei Luojia Laboratory,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Wuhan University,NERCMS, School of Computer Science, Hubei Luojia Laboratory,China","institution_ids":["https://openalex.org/I37461747"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102859655","display_name":"Yuhong Yang","orcid":"https://orcid.org/0000-0003-3001-7957"},"institutions":[{"id":"https://openalex.org/I37461747","display_name":"Wuhan University","ror":"https://ror.org/033vjfk17","country_code":"CN","type":"education","lineage":["https://openalex.org/I37461747"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuhong Yang","raw_affiliation_strings":["Wuhan University,NERCMS, School of Computer Science, Hubei Luojia Laboratory,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Wuhan University,NERCMS, School of Computer Science, Hubei Luojia Laboratory,China","institution_ids":["https://openalex.org/I37461747"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050125868","display_name":"Xinmeng Xu","orcid":"https://orcid.org/0009-0008-1607-7180"},"institutions":[{"id":"https://openalex.org/I37461747","display_name":"Wuhan University","ror":"https://ror.org/033vjfk17","country_code":"CN","type":"education","lineage":["https://openalex.org/I37461747"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xinmeng Xu","raw_affiliation_strings":["Wuhan University,NERCMS, School of Computer Science, Hubei Luojia Laboratory,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Wuhan University,NERCMS, School of Computer Science, Hubei Luojia Laboratory,China","institution_ids":["https://openalex.org/I37461747"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101499462","display_name":"Yanzhen Ren","orcid":"https://orcid.org/0000-0003-0799-5082"},"institutions":[{"id":"https://openalex.org/I37461747","display_name":"Wuhan University","ror":"https://ror.org/033vjfk17","country_code":"CN","type":"education","lineage":["https://openalex.org/I37461747"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yanzhen Ren","raw_affiliation_strings":["Wuhan University,School of Cyber Science and Engineering,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Wuhan University,School of Cyber Science and Engineering,China","institution_ids":["https://openalex.org/I37461747"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100749286","display_name":"Rong Zhu","orcid":"https://orcid.org/0000-0002-0687-7343"},"institutions":[{"id":"https://openalex.org/I37461747","display_name":"Wuhan University","ror":"https://ror.org/033vjfk17","country_code":"CN","type":"education","lineage":["https://openalex.org/I37461747"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Rong Zhu","raw_affiliation_strings":["Wuhan University,NERCMS, School of Computer Science, Hubei Luojia Laboratory,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Wuhan University,NERCMS, School of Computer Science, Hubei Luojia Laboratory,China","institution_ids":["https://openalex.org/I37461747"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.9349,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.72296103,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":98},"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/T10688","display_name":"Image and Signal Denoising Methods","score":0.9797000288963318,"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"}},"topics":[{"id":"https://openalex.org/T10688","display_name":"Image and Signal Denoising Methods","score":0.9797000288963318,"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/T10320","display_name":"Neural Networks and Applications","score":0.9765999913215637,"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/T11034","display_name":"Digital Filter Design and Implementation","score":0.9478999972343445,"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/computer-science","display_name":"Computer science","score":0.6952050924301147},{"id":"https://openalex.org/keywords/codec","display_name":"Codec","score":0.6657121777534485},{"id":"https://openalex.org/keywords/weighting","display_name":"Weighting","score":0.6336885690689087},{"id":"https://openalex.org/keywords/conditional-entropy","display_name":"Conditional entropy","score":0.5488525032997131},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.5443463325500488},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5155777335166931},{"id":"https://openalex.org/keywords/entropy","display_name":"Entropy (arrow of time)","score":0.4937911927700043},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3536075949668884},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.2237132489681244},{"id":"https://openalex.org/keywords/principle-of-maximum-entropy","display_name":"Principle of maximum entropy","score":0.11915743350982666},{"id":"https://openalex.org/keywords/telecommunications","display_name":"Telecommunications","score":0.10503894090652466},{"id":"https://openalex.org/keywords/acoustics","display_name":"Acoustics","score":0.09438541531562805}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6952050924301147},{"id":"https://openalex.org/C161765866","wikidata":"https://www.wikidata.org/wiki/Q184748","display_name":"Codec","level":2,"score":0.6657121777534485},{"id":"https://openalex.org/C183115368","wikidata":"https://www.wikidata.org/wiki/Q856577","display_name":"Weighting","level":2,"score":0.6336885690689087},{"id":"https://openalex.org/C101721835","wikidata":"https://www.wikidata.org/wiki/Q813908","display_name":"Conditional entropy","level":3,"score":0.5488525032997131},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.5443463325500488},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5155777335166931},{"id":"https://openalex.org/C106301342","wikidata":"https://www.wikidata.org/wiki/Q4117933","display_name":"Entropy (arrow of time)","level":2,"score":0.4937911927700043},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3536075949668884},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2237132489681244},{"id":"https://openalex.org/C9679016","wikidata":"https://www.wikidata.org/wiki/Q1417473","display_name":"Principle of maximum entropy","level":2,"score":0.11915743350982666},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.10503894090652466},{"id":"https://openalex.org/C24890656","wikidata":"https://www.wikidata.org/wiki/Q82811","display_name":"Acoustics","level":1,"score":0.09438541531562805},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","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/icassp49660.2025.10889560","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10889560","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320330944","display_name":"Nature","ror":null}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":27,"referenced_works":["https://openalex.org/W1567520911","https://openalex.org/W1995875735","https://openalex.org/W2130158387","https://openalex.org/W2159644785","https://openalex.org/W2168098717","https://openalex.org/W2752796333","https://openalex.org/W2972359262","https://openalex.org/W3110277971","https://openalex.org/W3214758449","https://openalex.org/W3215615641","https://openalex.org/W4307323391","https://openalex.org/W4372270198","https://openalex.org/W4375869380","https://openalex.org/W4386075800","https://openalex.org/W4386139314","https://openalex.org/W4392903089","https://openalex.org/W4392903389","https://openalex.org/W4392903887","https://openalex.org/W4402670057","https://openalex.org/W6633114069","https://openalex.org/W6639363673","https://openalex.org/W6739901393","https://openalex.org/W6744762798","https://openalex.org/W6750665317","https://openalex.org/W6777776875","https://openalex.org/W6853515095","https://openalex.org/W6948682473"],"related_works":["https://openalex.org/W4391590134","https://openalex.org/W2590770961","https://openalex.org/W3046039077","https://openalex.org/W1550757207","https://openalex.org/W2003770568","https://openalex.org/W2186435021","https://openalex.org/W2161963661","https://openalex.org/W2350489049","https://openalex.org/W2092738042","https://openalex.org/W1795191888"],"abstract_inverted_index":{"Existing":[0],"end-to-end":[1],"neural":[2],"codecs":[3,27],"have":[4,32],"made":[5],"great":[6],"progress":[7],"in":[8,19,188],"preserving":[9,192],"audio":[10,38,176,197],"quality.":[11,194],"Despite":[12],"their":[13,125],"success,":[14],"they":[15,68],"still":[16],"face":[17],"challenges":[18],"achieving":[20,174],"accurate":[21,119],"and":[22,100,123,148,191],"efficient":[23],"quantization.":[24],"Specifically,":[25],"these":[26,88],"often":[28],"overlook":[29],"which":[30,78,93],"features":[31,114,122],"a":[33,44,101],"greater":[34],"impact":[35],"on":[36,162],"perceptual":[37,193],"quality":[39,177],"during":[40],"quantization,":[41,116,129],"leading":[42],"to":[43,50,62,74],"quantization":[45,76,81,126,158],"error":[46],"distribution":[47],"that":[48,167],"fails":[49],"reflect":[51],"the":[52,141,144,153,157,163],"actual":[53],"importance":[54],"of":[55,112],"latent":[56,113,147],"features.":[57],"They":[58],"are":[59],"also":[60],"sensitive":[61],"unusual":[63],"data":[64],"points":[65],"(outliers)":[66],"because":[67],"use":[69],"Mean":[70],"Squared":[71],"Error":[72],"(MSE)":[73],"measure":[75],"errors,":[77],"can":[79,199],"increase":[80],"noise":[82],"or":[83],"spectral":[84],"artifacts.":[85],"To":[86],"address":[87],"limitations,":[89],"we":[90],"propose":[91],"AW-CEQCodec":[92],"integrates":[94],"an":[95],"Attention":[96],"Weighting":[97],"(AW)":[98],"module":[99],"Conditional":[102],"Entropy-driven":[103],"Quantization":[104],"(CEQ)":[105],"loss.":[106],"The":[107,195],"AW":[108],"enhances":[109],"key":[110],"regions":[111],"before":[115],"enabling":[117],"more":[118],"quantizing":[120],"critical":[121],"reducing":[124],"errors.":[127],"After":[128],"it":[130],"restores":[131],"global":[132],"details":[133],"from":[134],"dequantized":[135],"features,":[136,150],"improving":[137],"overall":[138],"reconstruction.":[139],"Moreover,":[140],"CEQ":[142],"minimizes":[143],"uncertainty":[145],"between":[146],"quantized":[149],"effectively":[151],"reflecting":[152],"distortion":[154,190],"introduced":[155],"by":[156],"module.":[159],"Experimental":[160],"results":[161],"CodecSuperb-STL":[164],"dataset":[165],"demonstrate":[166],"our":[168],"method":[169],"consistently":[170],"outperforms":[171],"baseline":[172],"approaches,":[173],"superior":[175],"at":[178,202],"bitrates":[179],"as":[180,182],"low":[181],"0.5":[183],"kbps,":[184],"confirming":[185],"its":[186],"effectiveness":[187],"minimizing":[189],"reconstruction":[196],"samples":[198],"be":[200],"find":[201],"https://huazhi1024.github.io/first-page.":[203]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}