{"id":"https://openalex.org/W3178321840","doi":"https://doi.org/10.1109/taslp.2021.3129994","title":"SoundStream: An End-to-End Neural Audio Codec","display_name":"SoundStream: An End-to-End Neural Audio Codec","publication_year":2021,"publication_date":"2021-11-23","ids":{"openalex":"https://openalex.org/W3178321840","doi":"https://doi.org/10.1109/taslp.2021.3129994","mag":"3178321840"},"language":"en","primary_location":{"id":"doi:10.1109/taslp.2021.3129994","is_oa":false,"landing_page_url":"https://doi.org/10.1109/taslp.2021.3129994","pdf_url":null,"source":{"id":"https://openalex.org/S4210169297","display_name":"IEEE/ACM Transactions on Audio Speech and Language Processing","issn_l":"2329-9290","issn":["2329-9290","2329-9304"],"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/ACM Transactions on Audio, Speech, and Language Processing","raw_type":"journal-article"},"type":"preprint","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2107.03312","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5047639590","display_name":"Neil Zeghidour","orcid":"https://orcid.org/0000-0001-6896-3987"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Neil Zeghidour","raw_affiliation_strings":["Research Google Inc., Mountain View, CA, USA","[Research, Google Inc, 93176 Mountain View, California, United States, 94043 (e-mail: neilz@google.com)]"],"raw_orcid":"https://orcid.org/0000-0001-6896-3987","affiliations":[{"raw_affiliation_string":"Research Google Inc., Mountain View, CA, USA","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"[Research, Google Inc, 93176 Mountain View, California, United States, 94043 (e-mail: neilz@google.com)]","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030815835","display_name":"Alejandro Luebs","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Alejandro Luebs","raw_affiliation_strings":["Research Google Inc., Mountain View, CA, USA","[Research, Google Inc, 93176 Mountain View, California, United States, (e-mail: aluebs@google.com)]"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Research Google Inc., Mountain View, CA, USA","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"[Research, Google Inc, 93176 Mountain View, California, United States, (e-mail: aluebs@google.com)]","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5073564704","display_name":"Ahmed S. Omran","orcid":"https://orcid.org/0000-0002-9145-7625"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ahmed Omran","raw_affiliation_strings":["Research Google Inc., Mountain View, CA, USA","[Research, Google Inc, 93176 Mountain View, California, United States, (e-mail: ahmedomran@google.com)]"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Research Google Inc., Mountain View, CA, USA","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"[Research, Google Inc, 93176 Mountain View, California, United States, (e-mail: ahmedomran@google.com)]","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5070792118","display_name":"Jan Skoglund","orcid":"https://orcid.org/0009-0008-0167-4628"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jan Skoglund","raw_affiliation_strings":["Research Google Inc., Mountain View, CA, USA","[Google, Inc., Mountain View, California, United States, 94043 (e-mail: jks@google.com)]"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Research Google Inc., Mountain View, CA, USA","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"[Google, Inc., Mountain View, California, United States, 94043 (e-mail: jks@google.com)]","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5033341878","display_name":"Marco Tagliasacchi","orcid":"https://orcid.org/0000-0002-7682-6795"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]},{"id":"https://openalex.org/I4210100430","display_name":"Google (Switzerland)","ror":"https://ror.org/014f9c269","country_code":"CH","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210100430","https://openalex.org/I4210128969"]}],"countries":["CH","US"],"is_corresponding":false,"raw_author_name":"Marco Tagliasacchi","raw_affiliation_strings":["Research Google Inc., Mountain View, CA, USA","[Research, Google Inc Zurich, Zurich, Switzerland, 8002 (e-mail: mtagliasacchi@google.com)]"],"raw_orcid":"https://orcid.org/0000-0002-7682-6795","affiliations":[{"raw_affiliation_string":"Research Google Inc., Mountain View, CA, USA","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"[Research, Google Inc Zurich, Zurich, Switzerland, 8002 (e-mail: mtagliasacchi@google.com)]","institution_ids":["https://openalex.org/I4210100430"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.8983,"has_fulltext":true,"cited_by_count":6,"citation_normalized_percentile":{"value":0.73406548,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":93,"max":96},"biblio":{"volume":"30","issue":null,"first_page":"495","last_page":"507"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":0.9998999834060669,"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":0.9998999834060669,"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/T11309","display_name":"Music and Audio Processing","score":0.9991000294685364,"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":0.9990000128746033,"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":0.8390601873397827},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.7157384157180786},{"id":"https://openalex.org/keywords/codec","display_name":"Codec","score":0.7081016302108765},{"id":"https://openalex.org/keywords/code-excited-linear-prediction","display_name":"Code-excited linear prediction","score":0.6297459006309509},{"id":"https://openalex.org/keywords/speech-coding","display_name":"Speech coding","score":0.6182061433792114},{"id":"https://openalex.org/keywords/adaptive-multi-rate-audio-codec","display_name":"Adaptive Multi-Rate audio codec","score":0.5648295283317566},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.5291588306427002},{"id":"https://openalex.org/keywords/speech-enhancement","display_name":"Speech enhancement","score":0.4733933210372925},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.4635092318058014},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.4616140127182007},{"id":"https://openalex.org/keywords/end-to-end-principle","display_name":"End-to-end principle","score":0.43202149868011475},{"id":"https://openalex.org/keywords/voice-activity-detection","display_name":"Voice activity detection","score":0.3639815151691437},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3205416798591614},{"id":"https://openalex.org/keywords/speech-processing","display_name":"Speech processing","score":0.2665785551071167},{"id":"https://openalex.org/keywords/noise-reduction","display_name":"Noise reduction","score":0.16380226612091064},{"id":"https://openalex.org/keywords/linear-predictive-coding","display_name":"Linear predictive coding","score":0.16173988580703735},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.1508747637271881}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8390601873397827},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.7157384157180786},{"id":"https://openalex.org/C161765866","wikidata":"https://www.wikidata.org/wiki/Q184748","display_name":"Codec","level":2,"score":0.7081016302108765},{"id":"https://openalex.org/C105964291","wikidata":"https://www.wikidata.org/wiki/Q856184","display_name":"Code-excited linear prediction","level":4,"score":0.6297459006309509},{"id":"https://openalex.org/C13895895","wikidata":"https://www.wikidata.org/wiki/Q3270773","display_name":"Speech coding","level":2,"score":0.6182061433792114},{"id":"https://openalex.org/C177067256","wikidata":"https://www.wikidata.org/wiki/Q4676210","display_name":"Adaptive Multi-Rate audio codec","level":4,"score":0.5648295283317566},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.5291588306427002},{"id":"https://openalex.org/C2776182073","wikidata":"https://www.wikidata.org/wiki/Q7575395","display_name":"Speech enhancement","level":3,"score":0.4733933210372925},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.4635092318058014},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.4616140127182007},{"id":"https://openalex.org/C74296488","wikidata":"https://www.wikidata.org/wiki/Q2527392","display_name":"End-to-end principle","level":2,"score":0.43202149868011475},{"id":"https://openalex.org/C204201278","wikidata":"https://www.wikidata.org/wiki/Q1332614","display_name":"Voice activity detection","level":3,"score":0.3639815151691437},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3205416798591614},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.2665785551071167},{"id":"https://openalex.org/C163294075","wikidata":"https://www.wikidata.org/wiki/Q581861","display_name":"Noise reduction","level":2,"score":0.16380226612091064},{"id":"https://openalex.org/C59883199","wikidata":"https://www.wikidata.org/wiki/Q1826438","display_name":"Linear predictive coding","level":3,"score":0.16173988580703735},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.1508747637271881},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.1109/taslp.2021.3129994","is_oa":false,"landing_page_url":"https://doi.org/10.1109/taslp.2021.3129994","pdf_url":null,"source":{"id":"https://openalex.org/S4210169297","display_name":"IEEE/ACM Transactions on Audio Speech and Language Processing","issn_l":"2329-9290","issn":["2329-9290","2329-9304"],"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/ACM Transactions on Audio, Speech, and Language Processing","raw_type":"journal-article"},{"id":"pmh:oai:arXiv.org:2107.03312","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2107.03312","pdf_url":"https://arxiv.org/pdf/2107.03312","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},{"id":"doi:10.48550/arxiv.2107.03312","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2107.03312","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"},{"id":"mag:3178321840","is_oa":false,"landing_page_url":null,"pdf_url":null,"source":null,"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":null}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2107.03312","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2107.03312","pdf_url":"https://arxiv.org/pdf/2107.03312","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},"sustainable_development_goals":[{"display_name":"Peace, Justice and strong institutions","score":0.550000011920929,"id":"https://metadata.un.org/sdg/16"}],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3178321840.pdf","grobid_xml":"https://content.openalex.org/works/W3178321840.grobid-xml"},"referenced_works_count":92,"referenced_works":["https://openalex.org/W1481955708","https://openalex.org/W1552314771","https://openalex.org/W1607435270","https://openalex.org/W1700083040","https://openalex.org/W1885680957","https://openalex.org/W1973681148","https://openalex.org/W2002182716","https://openalex.org/W2095705004","https://openalex.org/W2127218421","https://openalex.org/W2127870748","https://openalex.org/W2129913307","https://openalex.org/W2131738223","https://openalex.org/W2134383396","https://openalex.org/W2148371116","https://openalex.org/W2150593711","https://openalex.org/W2151626637","https://openalex.org/W2165291881","https://openalex.org/W2176412452","https://openalex.org/W2286601668","https://openalex.org/W2296581541","https://openalex.org/W2609317876","https://openalex.org/W2752796333","https://openalex.org/W2760103357","https://openalex.org/W2768814045","https://openalex.org/W2770119437","https://openalex.org/W2775336875","https://openalex.org/W2788851830","https://openalex.org/W2802034954","https://openalex.org/W2810843531","https://openalex.org/W2811079561","https://openalex.org/W2890983311","https://openalex.org/W2913399920","https://openalex.org/W2935711438","https://openalex.org/W2947590261","https://openalex.org/W2949382160","https://openalex.org/W2950237361","https://openalex.org/W2953331651","https://openalex.org/W2963091184","https://openalex.org/W2963103134","https://openalex.org/W2963182577","https://openalex.org/W2963321191","https://openalex.org/W2963341071","https://openalex.org/W2963782041","https://openalex.org/W2963799213","https://openalex.org/W2963921132","https://openalex.org/W2970006822","https://openalex.org/W2972354707","https://openalex.org/W2972359262","https://openalex.org/W2972443522","https://openalex.org/W2998572311","https://openalex.org/W3000389243","https://openalex.org/W3015780049","https://openalex.org/W3021164770","https://openalex.org/W3036601975","https://openalex.org/W3036682213","https://openalex.org/W3037038648","https://openalex.org/W3046970875","https://openalex.org/W3092028330","https://openalex.org/W3095095816","https://openalex.org/W3099782249","https://openalex.org/W3130248090","https://openalex.org/W3140429000","https://openalex.org/W3160576174","https://openalex.org/W3163243746","https://openalex.org/W3163662330","https://openalex.org/W3169418678","https://openalex.org/W3186609711","https://openalex.org/W4205169642","https://openalex.org/W4244017338","https://openalex.org/W6631190155","https://openalex.org/W6635645706","https://openalex.org/W6636170946","https://openalex.org/W6639363673","https://openalex.org/W6674330103","https://openalex.org/W6678914141","https://openalex.org/W6678969435","https://openalex.org/W6685562342","https://openalex.org/W6732429163","https://openalex.org/W6746278845","https://openalex.org/W6748409065","https://openalex.org/W6753018729","https://openalex.org/W6762931180","https://openalex.org/W6767111847","https://openalex.org/W6769196770","https://openalex.org/W6771763809","https://openalex.org/W6772349387","https://openalex.org/W6776218486","https://openalex.org/W6780218876","https://openalex.org/W6780365925","https://openalex.org/W6781251213","https://openalex.org/W6783867762","https://openalex.org/W6843673214"],"related_works":["https://openalex.org/W2100860879","https://openalex.org/W2963182577","https://openalex.org/W129817288","https://openalex.org/W2069969665","https://openalex.org/W1672246675","https://openalex.org/W1954114910","https://openalex.org/W2845558680","https://openalex.org/W2972354707","https://openalex.org/W2288779434","https://openalex.org/W2960731461","https://openalex.org/W2054812163","https://openalex.org/W2775749974","https://openalex.org/W2279551481","https://openalex.org/W2524123961","https://openalex.org/W1907552535","https://openalex.org/W2116603632","https://openalex.org/W2151549934","https://openalex.org/W2514199377","https://openalex.org/W1499890800","https://openalex.org/W1485475391"],"abstract_inverted_index":{"We":[0],"present":[1],"<italic":[2,27,148],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[3,28,149],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">SoundStream</i>":[4,29,150],",":[5],"a":[6,32,37,43,87,102,121,135],"novel":[7],"neural":[8],"audio":[9,19,73,142],"codec":[10],"that":[11],"can":[12,90],"efficiently":[13],"compress":[14],"speech,":[15],"music":[16],"and":[17,42,58,64,129,159,173],"general":[18],"at":[20,111,143,151,156,162,176,180],"bitrates":[21,94],"normally":[22],"targeted":[23],"by":[24,36],"speech-tailored":[25],"codecs.":[26],"relies":[30],"on":[31,134],"model":[33,89,117],"architecture":[34],"composed":[35],"fully":[38],"convolutional":[39],"encoder/decoder":[40],"network":[41],"residual":[44],"vector":[45],"quantizer,":[46],"which":[47,61,125,188],"are":[48,167],"trained":[49,110],"jointly":[50],"end-to-end.":[51],"Training":[52],"leverages":[53],"recent":[54],"advances":[55],"in":[56,131],"text-to-speech":[57],"speech":[59],"enhancement,":[60],"combine":[62],"adversarial":[63],"reconstruction":[65],"losses":[66],"to":[67,84,98,120,169],"allow":[68],"the":[69,116,177,181],"generation":[70],"of":[71],"high-quality":[72],"content":[74],"from":[75,95],"quantized":[76],"embeddings.":[77],"By":[78],"training":[79],"with":[80,101,108,184],"structured":[81],"dropout":[82],"applied":[83],"quantizer":[85],"layers,":[86],"single":[88],"operate":[91],"across":[92],"variable":[93],"3":[96,152],"kbps":[97,153,158],"18":[99],"kbps,":[100],"negligible":[103],"quality":[104],"loss":[105],"when":[106],"compared":[107],"models":[109],"fixed":[112],"bitrates.":[113],"In":[114,138],"addition,":[115],"is":[118],"amenable":[119],"low":[122],"latency":[123],"implementation,":[124],"supports":[126],"streamable":[127],"inference":[128],"runs":[130],"real":[132],"time":[133],"smartphone":[136],"CPU.":[137],"subjective":[139],"evaluations":[140],"using":[141],"24":[144],"kHz":[145],"sampling":[146],"rate,":[147],"outperforms":[154],"Opus":[155],"12":[157],"approaches":[160],"EVS":[161],"9.6":[163],"kbps.":[164],"Moreover,":[165],"we":[166,189],"able":[168],"perform":[170],"joint":[171],"compression":[172],"enhancement":[174],"either":[175],"encoder":[178],"or":[179],"decoder":[182],"side":[183],"no":[185],"additional":[186],"latency,":[187],"demonstrate":[190],"through":[191],"background":[192],"noise":[193],"suppression":[194],"for":[195],"speech.":[196]},"counts_by_year":[{"year":2023,"cited_by_count":2},{"year":2022,"cited_by_count":2},{"year":2021,"cited_by_count":2}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}