{"id":"https://openalex.org/W4224931676","doi":"https://doi.org/10.1109/icassp43922.2022.9746690","title":"Infergrad: Improving Diffusion Models for Vocoder by Considering Inference in Training","display_name":"Infergrad: Improving Diffusion Models for Vocoder by Considering Inference in Training","publication_year":2022,"publication_date":"2022-04-27","ids":{"openalex":"https://openalex.org/W4224931676","doi":"https://doi.org/10.1109/icassp43922.2022.9746690"},"language":"en","primary_location":{"id":"doi:10.1109/icassp43922.2022.9746690","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp43922.2022.9746690","pdf_url":null,"source":{"id":"https://openalex.org/S4363607702","display_name":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2022 - 2022 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/A5100718751","display_name":"Zehua Chen","orcid":"https://orcid.org/0000-0003-2127-7829"},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]},{"id":"https://openalex.org/I47508984","display_name":"Imperial College London","ror":"https://ror.org/041kmwe10","country_code":"GB","type":"education","lineage":["https://openalex.org/I47508984"]}],"countries":["CN","GB"],"is_corresponding":true,"raw_author_name":"Zehua Chen","raw_affiliation_strings":["Imperial College London,Department of EEE,London,UK,SW7 2AZ","Microsoft Azure Speech, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Imperial College London,Department of EEE,London,UK,SW7 2AZ","institution_ids":["https://openalex.org/I47508984"]},{"raw_affiliation_string":"Microsoft Azure Speech, Beijing, China","institution_ids":["https://openalex.org/I4210113369"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101522530","display_name":"Xu Tan","orcid":"https://orcid.org/0000-0001-5631-0639"},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xu Tan","raw_affiliation_strings":["Microsoft Research Asia,Beijing,China,100080"],"affiliations":[{"raw_affiliation_string":"Microsoft Research Asia,Beijing,China,100080","institution_ids":["https://openalex.org/I4210113369"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100360076","display_name":"Ke Wang","orcid":"https://orcid.org/0000-0001-5110-6591"},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ke Wang","raw_affiliation_strings":["Microsoft Azure Speech,Beijing,China,100080"],"affiliations":[{"raw_affiliation_string":"Microsoft Azure Speech,Beijing,China,100080","institution_ids":["https://openalex.org/I4210113369"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103063104","display_name":"Shifeng Pan","orcid":"https://orcid.org/0000-0002-9338-7800"},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shifeng Pan","raw_affiliation_strings":["Microsoft Azure Speech,Beijing,China,100080"],"affiliations":[{"raw_affiliation_string":"Microsoft Azure Speech,Beijing,China,100080","institution_ids":["https://openalex.org/I4210113369"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103001848","display_name":"Danilo P. Mandic","orcid":"https://orcid.org/0000-0001-8432-3963"},"institutions":[{"id":"https://openalex.org/I47508984","display_name":"Imperial College London","ror":"https://ror.org/041kmwe10","country_code":"GB","type":"education","lineage":["https://openalex.org/I47508984"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Danilo Mandic","raw_affiliation_strings":["Imperial College London,Department of EEE,London,UK,SW7 2AZ"],"affiliations":[{"raw_affiliation_string":"Imperial College London,Department of EEE,London,UK,SW7 2AZ","institution_ids":["https://openalex.org/I47508984"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5061724331","display_name":"Lei He","orcid":"https://orcid.org/0000-0003-3028-6305"},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lei He","raw_affiliation_strings":["Microsoft Azure Speech,Beijing,China,100080"],"affiliations":[{"raw_affiliation_string":"Microsoft Azure Speech,Beijing,China,100080","institution_ids":["https://openalex.org/I4210113369"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100329353","display_name":"Sheng Zhao","orcid":"https://orcid.org/0000-0002-9624-5381"},"institutions":[{"id":"https://openalex.org/I4210113369","display_name":"Microsoft Research Asia (China)","ror":"https://ror.org/0300m5276","country_code":"CN","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210113369"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Sheng Zhao","raw_affiliation_strings":["Microsoft Azure Speech,Beijing,China,100080"],"affiliations":[{"raw_affiliation_string":"Microsoft Azure Speech,Beijing,China,100080","institution_ids":["https://openalex.org/I4210113369"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5100718751"],"corresponding_institution_ids":["https://openalex.org/I4210113369","https://openalex.org/I47508984"],"apc_list":null,"apc_paid":null,"fwci":2.335,"has_fulltext":false,"cited_by_count":19,"citation_normalized_percentile":{"value":0.9093207,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"8432","last_page":"8436"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11309","display_name":"Music and Audio Processing","score":0.9997000098228455,"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":0.9997000098228455,"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":0.9994000196456909,"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":0.9994000196456909,"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/inference","display_name":"Inference","score":0.8829023838043213},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8157614469528198},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.7508636713027954},{"id":"https://openalex.org/keywords/frequentist-inference","display_name":"Frequentist inference","score":0.5147402286529541},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.48375821113586426},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.45008838176727295},{"id":"https://openalex.org/keywords/schedule","display_name":"Schedule","score":0.43079033493995667},{"id":"https://openalex.org/keywords/quality","display_name":"Quality (philosophy)","score":0.41861987113952637},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.41059955954551697},{"id":"https://openalex.org/keywords/bayesian-inference","display_name":"Bayesian inference","score":0.36413389444351196},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.34319964051246643},{"id":"https://openalex.org/keywords/bayesian-probability","display_name":"Bayesian probability","score":0.1638292670249939}],"concepts":[{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.8829023838043213},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8157614469528198},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.7508636713027954},{"id":"https://openalex.org/C162376815","wikidata":"https://www.wikidata.org/wiki/Q2158281","display_name":"Frequentist inference","level":4,"score":0.5147402286529541},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.48375821113586426},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.45008838176727295},{"id":"https://openalex.org/C68387754","wikidata":"https://www.wikidata.org/wiki/Q7271585","display_name":"Schedule","level":2,"score":0.43079033493995667},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.41861987113952637},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.41059955954551697},{"id":"https://openalex.org/C160234255","wikidata":"https://www.wikidata.org/wiki/Q812535","display_name":"Bayesian inference","level":3,"score":0.36413389444351196},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.34319964051246643},{"id":"https://openalex.org/C107673813","wikidata":"https://www.wikidata.org/wiki/Q812534","display_name":"Bayesian probability","level":2,"score":0.1638292670249939},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C111472728","wikidata":"https://www.wikidata.org/wiki/Q9471","display_name":"Epistemology","level":1,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp43922.2022.9746690","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp43922.2022.9746690","pdf_url":null,"source":{"id":"https://openalex.org/S4363607702","display_name":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":52,"referenced_works":["https://openalex.org/W1522301498","https://openalex.org/W1552314771","https://openalex.org/W1901129140","https://openalex.org/W2067295501","https://openalex.org/W2519091744","https://openalex.org/W2584032004","https://openalex.org/W2959300817","https://openalex.org/W2963090522","https://openalex.org/W2970006822","https://openalex.org/W2996286887","https://openalex.org/W3015338123","https://openalex.org/W3036167779","https://openalex.org/W3038172701","https://openalex.org/W3092028330","https://openalex.org/W3110257065","https://openalex.org/W3121370741","https://openalex.org/W3123097577","https://openalex.org/W3129651364","https://openalex.org/W3140633421","https://openalex.org/W3162926177","https://openalex.org/W3172148458","https://openalex.org/W3172617364","https://openalex.org/W3174758275","https://openalex.org/W3177150392","https://openalex.org/W3193411928","https://openalex.org/W4289761690","https://openalex.org/W4294619240","https://openalex.org/W4298580827","https://openalex.org/W6610566761","https://openalex.org/W6631190155","https://openalex.org/W6639824700","https://openalex.org/W6732429163","https://openalex.org/W6748409065","https://openalex.org/W6752910514","https://openalex.org/W6765775151","https://openalex.org/W6767111847","https://openalex.org/W6768435317","https://openalex.org/W6779823529","https://openalex.org/W6779871621","https://openalex.org/W6782760101","https://openalex.org/W6783182287","https://openalex.org/W6783713337","https://openalex.org/W6783867762","https://openalex.org/W6786375611","https://openalex.org/W6795261426","https://openalex.org/W6795288823","https://openalex.org/W6796730497","https://openalex.org/W6797095309","https://openalex.org/W6797906067","https://openalex.org/W6799642162","https://openalex.org/W6843673214","https://openalex.org/W6917585676"],"related_works":["https://openalex.org/W2097707447","https://openalex.org/W88131178","https://openalex.org/W2290754432","https://openalex.org/W2332232517","https://openalex.org/W232830309","https://openalex.org/W2372267530","https://openalex.org/W2969189870","https://openalex.org/W3015855446","https://openalex.org/W4303857162","https://openalex.org/W2965643117"],"abstract_inverted_index":{"Denoising":[0],"diffusion":[1,83],"probabilistic":[2],"models":[3,5],"(diffusion":[4],"for":[6,85,192,197],"short)":[7],"require":[8],"a":[9,47,82,114,121,126],"large":[10],"number":[11],"of":[12,43,145,153],"iterations":[13,49,97,191,196],"in":[14,32,57],"inference":[15,34,44,64,89,96,118,149],"to":[16,39,50,93,128],"achieve":[17],"the":[18,25,41,63,73,95,130,133,143,148,160,171,179,184],"generation":[19,59,101],"quality":[20,169,182],"that":[21,87,164],"matches":[22],"or":[23],"surpasses":[24],"state-of-the-art":[26],"generative":[27],"models,":[28],"which":[29],"invariably":[30],"results":[31,56],"slow":[33],"speed.":[35],"Previous":[36],"approaches":[37],"aim":[38],"optimize":[40],"choice":[42],"schedule":[45],"over":[46],"few":[48,122],"speed":[51],"up":[52],"inference.":[53],"However,":[54],"this":[55,77],"reduced":[58],"quality,":[60],"mainly":[61],"because":[62],"process":[65,90,116],"is":[66],"optimized":[67],"separately,":[68],"without":[69],"jointly":[70],"optimizing":[71],"with":[72,120,187],"training":[74,144],"process.":[75,150],"In":[76],"paper,":[78],"we":[79,107],"propose":[80],"InferGrad,":[81],"model":[84],"vocoder":[86],"incorporates":[88],"into":[91],"training,":[92,106],"reduce":[94],"while":[98,177],"maintaining":[99,178],"high":[100],"quality.":[102],"More":[103],"specifically,":[104],"during":[105],"generate":[108],"data":[109,137],"from":[110],"random":[111],"noise":[112],"through":[113,157],"reverse":[115],"under":[117,174],"schedules":[119],"iterations,":[123],"and":[124,135],"impose":[125],"loss":[127],"minimize":[129],"gap":[131],"between":[132],"generated":[134],"ground-truth":[136],"samples.":[138],"Then,":[139],"unlike":[140],"existing":[141],"approaches,":[142],"InferGrad":[146,154,165,193],"considers":[147],"The":[151],"advantages":[152],"are":[155],"demonstrated":[156],"experiments":[158],"on":[159],"LJSpeech":[161],"dataset":[162],"showing":[163],"achieves":[166],"better":[167],"voice":[168,181],"than":[170],"baseline":[172,185],"WaveGrad":[173],"same":[175,180],"conditions":[176],"as":[183],"but":[186],"3x":[188],"speedup":[189],"(2":[190],"vs":[194],"6":[195],"WaveGrad).":[198]},"counts_by_year":[{"year":2025,"cited_by_count":6},{"year":2024,"cited_by_count":5},{"year":2023,"cited_by_count":6},{"year":2022,"cited_by_count":2}],"updated_date":"2026-01-17T23:10:49.606395","created_date":"2025-10-10T00:00:00"}