{"id":"https://openalex.org/W4408355696","doi":"https://doi.org/10.1109/icassp49660.2025.10890341","title":"PTQ4ADM: Post-Training Quantization for Efficient Text Conditional Audio Diffusion Models","display_name":"PTQ4ADM: Post-Training Quantization for Efficient Text Conditional Audio Diffusion Models","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408355696","doi":"https://doi.org/10.1109/icassp49660.2025.10890341"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10890341","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10890341","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":"conference-paper","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/A5112360469","display_name":"Jayneel Vora","orcid":"https://orcid.org/0000-0002-0286-6380"},"institutions":[{"id":"https://openalex.org/I84218800","display_name":"University of California, Davis","ror":"https://ror.org/05rrcem69","country_code":"US","type":"education","lineage":["https://openalex.org/I84218800"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jayneel Vora","raw_affiliation_strings":["UC Davis,CA,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC Davis,CA,USA","institution_ids":["https://openalex.org/I84218800"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5021635802","display_name":"Aditya Krishnan","orcid":null},"institutions":[{"id":"https://openalex.org/I84218800","display_name":"University of California, Davis","ror":"https://ror.org/05rrcem69","country_code":"US","type":"education","lineage":["https://openalex.org/I84218800"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Aditya Krishnan","raw_affiliation_strings":["UC Davis,CA,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC Davis,CA,USA","institution_ids":["https://openalex.org/I84218800"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5019448790","display_name":"Nader Bouacida","orcid":"https://orcid.org/0000-0002-4571-8484"},"institutions":[{"id":"https://openalex.org/I84218800","display_name":"University of California, Davis","ror":"https://ror.org/05rrcem69","country_code":"US","type":"education","lineage":["https://openalex.org/I84218800"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Nader Bouacida","raw_affiliation_strings":["UC Davis,CA,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC Davis,CA,USA","institution_ids":["https://openalex.org/I84218800"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5012989760","display_name":"Prabhu Shankar","orcid":"https://orcid.org/0000-0001-9909-8527"},"institutions":[{"id":"https://openalex.org/I4210160856","display_name":"UC Davis Health","ror":"https://ror.org/05q8kyc69","country_code":"US","type":"healthcare","lineage":["https://openalex.org/I4210160856"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Prabhu R V Shankar","raw_affiliation_strings":["UC Davis Health,CA,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"UC Davis Health,CA,USA","institution_ids":["https://openalex.org/I4210160856"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5086439160","display_name":"Prasant Mohapatra","orcid":"https://orcid.org/0000-0002-2768-5308"},"institutions":[{"id":"https://openalex.org/I93320256","display_name":"University of Tampa","ror":"https://ror.org/007h1g065","country_code":"US","type":"education","lineage":["https://openalex.org/I93320256"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Prasant Mohapatra","raw_affiliation_strings":["USF Tampa,FL,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"USF Tampa,FL,USA","institution_ids":["https://openalex.org/I93320256"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.04585907,"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":0.9991999864578247,"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":0.9991999864578247,"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":0.9955999851226807,"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/T10860","display_name":"Speech and Audio Processing","score":0.9896000027656555,"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":0.740561842918396},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.6514763236045837},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.6110780835151672},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.48009803891181946},{"id":"https://openalex.org/keywords/speech-coding","display_name":"Speech coding","score":0.44566264748573303},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.38706615567207336},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.1964559257030487}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.740561842918396},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.6514763236045837},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.6110780835151672},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.48009803891181946},{"id":"https://openalex.org/C13895895","wikidata":"https://www.wikidata.org/wiki/Q3270773","display_name":"Speech coding","level":2,"score":0.44566264748573303},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.38706615567207336},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.1964559257030487},{"id":"https://openalex.org/C153294291","wikidata":"https://www.wikidata.org/wiki/Q25261","display_name":"Meteorology","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/icassp49660.2025.10890341","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10890341","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":[{"score":0.4099999964237213,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":25,"referenced_works":["https://openalex.org/W3134329288","https://openalex.org/W4285605725","https://openalex.org/W4312933868","https://openalex.org/W4367359628","https://openalex.org/W4381786045","https://openalex.org/W4386065704","https://openalex.org/W4387969125","https://openalex.org/W4390874074","https://openalex.org/W4401980590","https://openalex.org/W4402917085","https://openalex.org/W4403780831","https://openalex.org/W6679045638","https://openalex.org/W6779823529","https://openalex.org/W6793578827","https://openalex.org/W6796144397","https://openalex.org/W6845479124","https://openalex.org/W6849105126","https://openalex.org/W6849109464","https://openalex.org/W6849416043","https://openalex.org/W6852824296","https://openalex.org/W6852874946","https://openalex.org/W6852954323","https://openalex.org/W6853142820","https://openalex.org/W6860110139","https://openalex.org/W6861614106"],"related_works":["https://openalex.org/W230091440","https://openalex.org/W2233261550","https://openalex.org/W2810751659","https://openalex.org/W258997015","https://openalex.org/W2997094352","https://openalex.org/W1911859126","https://openalex.org/W2120730869","https://openalex.org/W1569721167","https://openalex.org/W2541680182","https://openalex.org/W2166699153"],"abstract_inverted_index":{"Denoising":[0],"diffusion":[1,61,93],"models":[2,62,139],"have":[3],"emerged":[4],"as":[5],"state-of-the-art":[6],"in":[7,77,168,176,204],"generative":[8],"tasks":[9,79],"across":[10,72],"image,":[11],"audio,":[12],"and":[13,19,33,105,124,137,186],"video":[14],"domains,":[15],"producing":[16],"high-quality,":[17],"diverse,":[18],"contextually":[20],"relevant":[21],"data.":[22],"However,":[23],"their":[24],"broader":[25],"adoption":[26],"is":[27],"limited":[28],"by":[29,48,154],"high":[30],"computational":[31],"costs":[32],"large":[34],"memory":[35],"footprints.":[36],"Post-training":[37],"quantization":[38,70],"(PTQ)":[39],"offers":[40],"a":[41,87,100],"promising":[42],"approach":[43,133],"to":[44,60,68,149,156,164,183],"mitigate":[45],"these":[46],"challenges":[47],"reducing":[49],"model":[50,152],"complexity":[51],"through":[52],"low-bandwidth":[53],"parameters.":[54],"Yet,":[55],"direct":[56],"application":[57],"of":[58,121,202],"PTQ":[59],"can":[63,180],"degrade":[64],"synthesis":[65,128,160],"quality":[66,161,191],"due":[67],"accumulated":[69],"noise":[71],"multiple":[73],"denoising":[74],"steps,":[75],"particularly":[76],"conditional":[78],"like":[80],"text-to-audio":[81],"synthesis.":[82],"This":[83,193],"work":[84,194],"introduces":[85],"PTQ4ADM,":[86],"novel":[88],"framework":[89],"for":[90,113,140,198],"quantizing":[91],"audio":[92,122,142],"models(ADMs).":[94],"Our":[95],"key":[96],"contributions":[97],"include":[98],"(1)":[99],"coverage-driven":[101],"prompt":[102],"augmentation":[103],"method":[104],"(2)":[106],"an":[107],"activation-aware":[108],"calibration":[109],"set":[110],"generation":[111],"algorithm":[112],"text-conditional":[114,141],"ADMs.":[115],"These":[116],"techniques":[117],"ensure":[118],"comprehensive":[119],"coverage":[120],"aspects":[123],"modalities":[125],"while":[126,158],"preserving":[127],"fidelity.":[129],"We":[130,171],"validate":[131],"our":[132],"on":[134],"TANGO,":[135],"Make-An-Audio,":[136],"AudioLDM":[138],"generation.":[143],"Extensive":[144],"experiments":[145],"demonstrate":[146],"PTQ4ADM\u2019s":[147],"capability":[148],"reduce":[150],"the":[151,177,196],"size":[153],"up":[155],"70%":[157],"achieving":[159],"metrics":[162],"comparable":[163],"full-precision":[165],"models(<5%":[166],"increase":[167],"FD":[169],"scores).":[170],"show":[172],"that":[173],"specific":[174],"layers":[175],"backbone":[178],"network":[179],"be":[181],"quantized":[182],"4-bit":[184],"weights":[185],"8-bit":[187],"activations":[188],"without":[189],"significant":[190],"loss.":[192],"paves":[195],"way":[197],"more":[199],"efficient":[200],"deployment":[201],"ADMs":[203],"resource-constrained":[205],"environments.":[206]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}