{"id":"https://openalex.org/W7123508531","doi":"https://doi.org/10.1109/mmsp64401.2025.11324336","title":"Lightweight DNN for Full-Band Speech Denoising on Mobile Devices: Exploiting Long and Short Temporal Patterns","display_name":"Lightweight DNN for Full-Band Speech Denoising on Mobile Devices: Exploiting Long and Short Temporal Patterns","publication_year":2025,"publication_date":"2025-09-21","ids":{"openalex":"https://openalex.org/W7123508531","doi":"https://doi.org/10.1109/mmsp64401.2025.11324336"},"language":null,"primary_location":{"id":"doi:10.1109/mmsp64401.2025.11324336","is_oa":false,"landing_page_url":"https://doi.org/10.1109/mmsp64401.2025.11324336","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Workshop on Multimedia Signal Processing (MMSP)","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/A5122976130","display_name":"Konstantinos Drossos","orcid":null},"institutions":[{"id":"https://openalex.org/I2738502077","display_name":"Nokia (Finland)","ror":"https://ror.org/04pkc8m17","country_code":"FI","type":"company","lineage":["https://openalex.org/I2738502077"]}],"countries":["FI"],"is_corresponding":true,"raw_author_name":"Konstantinos Drossos","raw_affiliation_strings":["Nokia Technologies,Espoo,Finland"],"affiliations":[{"raw_affiliation_string":"Nokia Technologies,Espoo,Finland","institution_ids":["https://openalex.org/I2738502077"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078854716","display_name":"Mikko Heikkinen","orcid":"https://orcid.org/0000-0001-9898-1916"},"institutions":[{"id":"https://openalex.org/I2738502077","display_name":"Nokia (Finland)","ror":"https://ror.org/04pkc8m17","country_code":"FI","type":"company","lineage":["https://openalex.org/I2738502077"]}],"countries":["FI"],"is_corresponding":false,"raw_author_name":"Mikko Heikkinen","raw_affiliation_strings":["Nokia Technologies,Tampere,Finland"],"affiliations":[{"raw_affiliation_string":"Nokia Technologies,Tampere,Finland","institution_ids":["https://openalex.org/I2738502077"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5058998799","display_name":"Paschalis Tsiaflakis","orcid":"https://orcid.org/0000-0001-6085-3740"},"institutions":[{"id":"https://openalex.org/I4210160875","display_name":"Nokia (Belgium)","ror":"https://ror.org/04xze1462","country_code":"BE","type":"company","lineage":["https://openalex.org/I2738502077","https://openalex.org/I4210160875"]}],"countries":["BE"],"is_corresponding":false,"raw_author_name":"Paschalis Tsiaflakis","raw_affiliation_strings":["Nokia Bell Labs,Antwerp,Belgium"],"affiliations":[{"raw_affiliation_string":"Nokia Bell Labs,Antwerp,Belgium","institution_ids":["https://openalex.org/I4210160875"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5122976130"],"corresponding_institution_ids":["https://openalex.org/I2738502077"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.69393722,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"168","last_page":"173"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":0.9007999897003174,"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":0.9007999897003174,"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":0.050700001418590546,"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/T10283","display_name":"Hearing Loss and Rehabilitation","score":0.01119999960064888,"subfield":{"id":"https://openalex.org/subfields/2805","display_name":"Cognitive Neuroscience"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/normalization","display_name":"Normalization (sociology)","score":0.6301000118255615},{"id":"https://openalex.org/keywords/noise-reduction","display_name":"Noise reduction","score":0.5547000169754028},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.42800000309944153},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.4180000126361847},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.41679999232292175},{"id":"https://openalex.org/keywords/recurrent-neural-network","display_name":"Recurrent neural network","score":0.3937000036239624},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.38589999079704285},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.38510000705718994}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8230999708175659},{"id":"https://openalex.org/C136886441","wikidata":"https://www.wikidata.org/wiki/Q926129","display_name":"Normalization (sociology)","level":2,"score":0.6301000118255615},{"id":"https://openalex.org/C163294075","wikidata":"https://www.wikidata.org/wiki/Q581861","display_name":"Noise reduction","level":2,"score":0.5547000169754028},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5295000076293945},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5033000111579895},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.42800000309944153},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.4180000126361847},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.41679999232292175},{"id":"https://openalex.org/C147168706","wikidata":"https://www.wikidata.org/wiki/Q1457734","display_name":"Recurrent neural network","level":3,"score":0.3937000036239624},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.38589999079704285},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.38510000705718994},{"id":"https://openalex.org/C2776182073","wikidata":"https://www.wikidata.org/wiki/Q7575395","display_name":"Speech enhancement","level":3,"score":0.3174000084400177},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.31380000710487366},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.2994999885559082},{"id":"https://openalex.org/C104267543","wikidata":"https://www.wikidata.org/wiki/Q208163","display_name":"Signal processing","level":3,"score":0.298799991607666},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.2976999878883362},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.29350000619888306},{"id":"https://openalex.org/C2779843651","wikidata":"https://www.wikidata.org/wiki/Q7390335","display_name":"SIGNAL (programming language)","level":2,"score":0.2896000146865845},{"id":"https://openalex.org/C29265498","wikidata":"https://www.wikidata.org/wiki/Q7047719","display_name":"Noise measurement","level":3,"score":0.2786000072956085},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.2761000096797943},{"id":"https://openalex.org/C142433447","wikidata":"https://www.wikidata.org/wiki/Q7806653","display_name":"Time\u2013frequency analysis","level":3,"score":0.2685000002384186},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.2651999890804291},{"id":"https://openalex.org/C46637626","wikidata":"https://www.wikidata.org/wiki/Q6693015","display_name":"Low latency (capital markets)","level":2,"score":0.26420000195503235},{"id":"https://openalex.org/C204201278","wikidata":"https://www.wikidata.org/wiki/Q1332614","display_name":"Voice activity detection","level":3,"score":0.26100000739097595},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2506999969482422}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/mmsp64401.2025.11324336","is_oa":false,"landing_page_url":"https://doi.org/10.1109/mmsp64401.2025.11324336","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Workshop on Multimedia Signal Processing (MMSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.5072733759880066,"display_name":"Decent work and economic growth","id":"https://metadata.un.org/sdg/8"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":20,"referenced_works":["https://openalex.org/W1552314771","https://openalex.org/W1901129140","https://openalex.org/W2067295501","https://openalex.org/W2603567530","https://openalex.org/W2952218014","https://openalex.org/W2963163009","https://openalex.org/W2963341071","https://openalex.org/W2963868408","https://openalex.org/W2964058413","https://openalex.org/W2982083293","https://openalex.org/W2998572311","https://openalex.org/W3095248373","https://openalex.org/W3161480375","https://openalex.org/W3162501355","https://openalex.org/W3205770165","https://openalex.org/W4225905067","https://openalex.org/W4312570973","https://openalex.org/W4392903568","https://openalex.org/W4392909774","https://openalex.org/W4401609277"],"related_works":[],"abstract_inverted_index":{"Speech":[0],"denoising":[1,126,178],"(SD)":[2],"is":[3,95,173],"an":[4,138,189],"important":[5],"task":[6],"of":[7,107],"many,":[8],"if":[9],"not":[10,54],"all,":[11],"modern":[12,167],"signal":[13,123],"processing":[14],"chains":[15],"used":[16],"in":[17,121,187],"devices":[18],"and":[19,28,79,89,110,117,124,156,180,196],"for":[20,35,40,51,83,114,153],"everyday-life":[21],"applications.":[22],"While":[23],"there":[24],"are":[25,38,53],"many":[26],"published":[27],"powerful":[29],"deep":[30],"neural":[31,112],"network":[32],"(DNN)-based":[33],"methods":[34,50],"SD,":[36,85],"few":[37],"optimized":[39],"resource-constrained":[41],"platforms":[42],"such":[43],"as":[44,137],"mobile":[45,168],"devices.":[46],"Additionally,":[47],"most":[48],"DNN-based":[49,81],"SD":[52,199],"focusing":[55],"on":[56,97,131,165],"full-band":[57,84],"(FB)":[58],"signals,":[59],"i.e.":[60],"having":[61],"48":[62],"kHz":[63],"sampling":[64],"rate,":[65],"and/or":[66],"low":[67,77,197],"latency":[68,198],"cases.":[69],"In":[70],"this":[71],"paper":[72],"we":[73],"present":[74],"a":[75,98,132,158,166],"causal,":[76],"latency,":[78],"lightweight":[80],"method":[82,94,172],"leveraging":[86],"both":[87],"short":[88,116],"long":[90,118],"temporal":[91,105,119],"patterns.":[92],"The":[93,128,170],"based":[96],"modified":[99],"UNet":[100],"architecture":[101],"employing":[102],"look-back":[103],"frames,":[104],"spanning":[106],"convolutional":[108],"kernels,":[109],"recurrent":[111],"networks":[113],"exploiting":[115],"patterns":[120],"the":[122,140],"estimated":[125],"mask.":[127],"DNN":[129],"operates":[130],"causal":[133,150],"frame-by-frame":[134],"basis":[135],"taking":[136],"input":[139],"STFT":[141],"magnitude,":[142],"utilizes":[143],"inverted":[144],"bottlenecks":[145],"inspired":[146],"by":[147],"MobileNet,":[148],"employs":[149],"instance":[151],"normalization":[152],"channel-wise":[154],"normalization,":[155],"achieves":[157],"real-time":[159],"factor":[160],"below":[161],"0.02":[162],"when":[163],"deployed":[164],"phone.":[169],"proposed":[171],"evaluated":[174],"using":[175],"established":[176],"speech":[177],"metrics":[179],"publicly":[181],"available":[182],"datasets,":[183],"demonstrating":[184],"its":[185],"effectiveness":[186],"achieving":[188],"(SI-)SDR":[190],"value":[191],"that":[192],"outperforms":[193],"existing":[194],"FB":[195],"methods.":[200]},"counts_by_year":[],"updated_date":"2026-01-14T23:44:37.837170","created_date":"2026-01-14T00:00:00"}