{"id":"https://openalex.org/W4408352160","doi":"https://doi.org/10.1109/icassp49660.2025.10888898","title":"FlowMAC: Conditional Flow Matching for Audio Coding at Low Bit Rates","display_name":"FlowMAC: Conditional Flow Matching for Audio Coding at Low Bit Rates","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408352160","doi":"https://doi.org/10.1109/icassp49660.2025.10888898"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10888898","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10888898","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/A5039716073","display_name":"Nicola Pia","orcid":"https://orcid.org/0000-0003-0987-863X"},"institutions":[{"id":"https://openalex.org/I4210124274","display_name":"Fraunhofer Institute for Integrated Circuits","ror":"https://ror.org/024ape423","country_code":"DE","type":"facility","lineage":["https://openalex.org/I4210124274","https://openalex.org/I4923324"]}],"countries":["DE"],"is_corresponding":true,"raw_author_name":"Nicola Pia","raw_affiliation_strings":["Fraunhofer IIS,Erlangen,Germany"],"affiliations":[{"raw_affiliation_string":"Fraunhofer IIS,Erlangen,Germany","institution_ids":["https://openalex.org/I4210124274"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5107196494","display_name":"Martin J. Strauss","orcid":null},"institutions":[{"id":"https://openalex.org/I4210123192","display_name":"International Audio Laboratories Erlangen","ror":"https://ror.org/02mkz3e80","country_code":"DE","type":"facility","lineage":["https://openalex.org/I181369854","https://openalex.org/I4210123192","https://openalex.org/I4210124274","https://openalex.org/I4923324"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Martin Strauss","raw_affiliation_strings":["International Audio Laboratories Erlangen,Erlangen,Germany"],"affiliations":[{"raw_affiliation_string":"International Audio Laboratories Erlangen,Erlangen,Germany","institution_ids":["https://openalex.org/I4210123192"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5009684245","display_name":"Markus Multrus","orcid":null},"institutions":[{"id":"https://openalex.org/I4210124274","display_name":"Fraunhofer Institute for Integrated Circuits","ror":"https://ror.org/024ape423","country_code":"DE","type":"facility","lineage":["https://openalex.org/I4210124274","https://openalex.org/I4923324"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Markus Multrus","raw_affiliation_strings":["Fraunhofer IIS,Erlangen,Germany"],"affiliations":[{"raw_affiliation_string":"Fraunhofer IIS,Erlangen,Germany","institution_ids":["https://openalex.org/I4210124274"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5022331149","display_name":"Bernd Edler","orcid":null},"institutions":[{"id":"https://openalex.org/I4210123192","display_name":"International Audio Laboratories Erlangen","ror":"https://ror.org/02mkz3e80","country_code":"DE","type":"facility","lineage":["https://openalex.org/I181369854","https://openalex.org/I4210123192","https://openalex.org/I4210124274","https://openalex.org/I4923324"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Bernd Edler","raw_affiliation_strings":["International Audio Laboratories Erlangen,Erlangen,Germany"],"affiliations":[{"raw_affiliation_string":"International Audio Laboratories Erlangen,Erlangen,Germany","institution_ids":["https://openalex.org/I4210123192"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5039716073"],"corresponding_institution_ids":["https://openalex.org/I4210124274"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.04191864,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"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/T11309","display_name":"Music and Audio Processing","score":0.9962000250816345,"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/T11309","display_name":"Music and Audio Processing","score":0.9962000250816345,"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/T10901","display_name":"Advanced Data Compression Techniques","score":0.9932000041007996,"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/T11233","display_name":"Advanced Adaptive Filtering Techniques","score":0.9914000034332275,"subfield":{"id":"https://openalex.org/subfields/2206","display_name":"Computational Mechanics"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7333906888961792},{"id":"https://openalex.org/keywords/coding","display_name":"Coding (social sciences)","score":0.5003974437713623},{"id":"https://openalex.org/keywords/speech-coding","display_name":"Speech coding","score":0.4859722852706909},{"id":"https://openalex.org/keywords/harmonic-vector-excitation-coding","display_name":"Harmonic Vector Excitation Coding","score":0.48116931319236755},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.4732910990715027},{"id":"https://openalex.org/keywords/bit-rate","display_name":"Bit rate","score":0.45970141887664795},{"id":"https://openalex.org/keywords/sub-band-coding","display_name":"Sub-band coding","score":0.44042155146598816},{"id":"https://openalex.org/keywords/matching","display_name":"Matching (statistics)","score":0.43578779697418213},{"id":"https://openalex.org/keywords/real-time-computing","display_name":"Real-time computing","score":0.19175490736961365},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.14560005068778992},{"id":"https://openalex.org/keywords/statistics","display_name":"Statistics","score":0.10441690683364868}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7333906888961792},{"id":"https://openalex.org/C179518139","wikidata":"https://www.wikidata.org/wiki/Q5140297","display_name":"Coding (social sciences)","level":2,"score":0.5003974437713623},{"id":"https://openalex.org/C13895895","wikidata":"https://www.wikidata.org/wiki/Q3270773","display_name":"Speech coding","level":2,"score":0.4859722852706909},{"id":"https://openalex.org/C80167644","wikidata":"https://www.wikidata.org/wiki/Q463990","display_name":"Harmonic Vector Excitation Coding","level":3,"score":0.48116931319236755},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.4732910990715027},{"id":"https://openalex.org/C2987812609","wikidata":"https://www.wikidata.org/wiki/Q194158","display_name":"Bit rate","level":2,"score":0.45970141887664795},{"id":"https://openalex.org/C98526533","wikidata":"https://www.wikidata.org/wiki/Q1691938","display_name":"Sub-band coding","level":3,"score":0.44042155146598816},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.43578779697418213},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.19175490736961365},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.14560005068778992},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.10441690683364868}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/icassp49660.2025.10888898","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10888898","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"},{"id":"pmh:oai:publica.fraunhofer.de:publica/512799","is_oa":false,"landing_page_url":"https://publica.fraunhofer.de/handle/publica/512799","pdf_url":null,"source":{"id":"https://openalex.org/S4306400318","display_name":"Fraunhofer-Publica (Fraunhofer-Gesellschaft)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I4923324","host_organization_name":"Fraunhofer-Gesellschaft","host_organization_lineage":["https://openalex.org/I4923324"],"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":"conference paper"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":29,"referenced_works":["https://openalex.org/W1590793133","https://openalex.org/W2129913307","https://openalex.org/W2752796333","https://openalex.org/W2775336875","https://openalex.org/W2963182577","https://openalex.org/W2972354707","https://openalex.org/W2972359262","https://openalex.org/W3015268401","https://openalex.org/W3048593389","https://openalex.org/W3188073270","https://openalex.org/W3215615641","https://openalex.org/W4225287045","https://openalex.org/W4284972581","https://openalex.org/W4307323391","https://openalex.org/W4385680913","https://openalex.org/W4392903389","https://openalex.org/W4392931276","https://openalex.org/W4392931975","https://openalex.org/W4404740148","https://openalex.org/W6760712927","https://openalex.org/W6779577414","https://openalex.org/W6795261426","https://openalex.org/W6838843145","https://openalex.org/W6840815571","https://openalex.org/W6846539466","https://openalex.org/W6853096648","https://openalex.org/W6853515095","https://openalex.org/W6853888607","https://openalex.org/W6862995742"],"related_works":["https://openalex.org/W2021359362","https://openalex.org/W1820989711","https://openalex.org/W1898356410","https://openalex.org/W1655538338","https://openalex.org/W1560214705","https://openalex.org/W1568040281","https://openalex.org/W2054918484","https://openalex.org/W2200246965","https://openalex.org/W3004303670","https://openalex.org/W1976151278"],"abstract_inverted_index":{"This":[0,54,117],"paper":[1],"introduces":[2],"FlowMAC,":[3],"a":[4,27,40,50,60,70,105],"novel":[5],"neural":[6,94],"audio":[7,12,67,95],"codec":[8],"for":[9],"high-quality":[10,51],"general":[11,66],"compression":[13],"at":[14,83,97],"low":[15],"bit":[16,100],"rates":[17],"based":[18],"on":[19,121],"conditional":[20],"flow":[21,43],"matching":[22],"(CFM).":[23],"FlowMAC":[24,82,103],"jointly":[25],"learns":[26],"mel":[28,52],"spectrogram":[29],"encoder,":[30],"quantizer":[31],"and":[32,73,92,115],"decoder.":[33],"At":[34],"inference":[35,107],"time":[36,58],"the":[37,56,99],"decoder":[38],"integrates":[39],"continuous":[41],"normalizing":[42],"via":[44],"an":[45],"ODE":[46],"solver":[47],"to":[48,65,111],"generate":[49],"spectrogram.":[53],"is":[55,63],"first":[57],"that":[59,81],"CFM-based":[61],"approach":[62],"applied":[64],"coding,":[68],"enabling":[69],"scalable,":[71],"simple":[72],"memory":[74],"efficient":[75],"training.":[76],"Our":[77],"subjective":[78],"evaluations":[79],"show":[80],"3":[84],"kbps":[85],"achieves":[86],"similar":[87],"quality":[88],"as":[89],"state-of-the-art":[90],"GAN-based":[91],"DDPM-based":[93],"codecs":[96],"double":[98],"rate.":[101],"Moreover,":[102],"offers":[104],"tunable":[106],"pipeline,":[108],"which":[109],"permits":[110],"trade":[112],"off":[113],"complexity":[114],"quality.":[116,127],"enables":[118],"real-time":[119],"coding":[120],"CPU,":[122],"while":[123],"maintaining":[124],"high":[125],"perceptual":[126]},"counts_by_year":[],"updated_date":"2026-03-28T08:17:26.163206","created_date":"2025-10-10T00:00:00"}