{"id":"https://openalex.org/W3160919572","doi":"https://doi.org/10.1109/icassp39728.2021.9414403","title":"Lightspeech: Lightweight and Fast Text to Speech with Neural Architecture Search","display_name":"Lightspeech: Lightweight and Fast Text to Speech with Neural Architecture Search","publication_year":2021,"publication_date":"2021-05-13","ids":{"openalex":"https://openalex.org/W3160919572","doi":"https://doi.org/10.1109/icassp39728.2021.9414403","mag":"3160919572"},"language":"en","primary_location":{"id":"doi:10.1109/icassp39728.2021.9414403","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9414403","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2021 - 2021 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/A5030131487","display_name":"Renqian Luo","orcid":"https://orcid.org/0000-0002-9062-3484"},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Renqian Luo","raw_affiliation_strings":["University of Science and Technology of China"],"affiliations":[{"raw_affiliation_string":"University of Science and Technology of China","institution_ids":["https://openalex.org/I126520041"]}]},{"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"],"affiliations":[{"raw_affiliation_string":"Microsoft Research Asia","institution_ids":["https://openalex.org/I4210113369"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100431163","display_name":"Rui Wang","orcid":"https://orcid.org/0000-0001-9048-2979"},"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":"Rui Wang","raw_affiliation_strings":["Microsoft Research Asia"],"affiliations":[{"raw_affiliation_string":"Microsoft Research Asia","institution_ids":["https://openalex.org/I4210113369"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020025718","display_name":"Tao Qin","orcid":"https://orcid.org/0000-0002-9095-0776"},"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":"Tao Qin","raw_affiliation_strings":["Microsoft Research Asia"],"affiliations":[{"raw_affiliation_string":"Microsoft Research Asia","institution_ids":["https://openalex.org/I4210113369"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041283996","display_name":"Jinzhu Li","orcid":"https://orcid.org/0000-0002-8835-8837"},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jinzhu Li","raw_affiliation_strings":["Microsoft Azure Speech"],"affiliations":[{"raw_affiliation_string":"Microsoft Azure Speech","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100329353","display_name":"Sheng Zhao","orcid":"https://orcid.org/0000-0002-9624-5381"},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Sheng Zhao","raw_affiliation_strings":["Microsoft Azure Speech"],"affiliations":[{"raw_affiliation_string":"Microsoft Azure Speech","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048237545","display_name":"Enhong Chen","orcid":"https://orcid.org/0000-0002-4835-4102"},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Enhong Chen","raw_affiliation_strings":["University of Science and Technology of China"],"affiliations":[{"raw_affiliation_string":"University of Science and Technology of China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101884287","display_name":"Tie\u2010Yan Liu","orcid":"https://orcid.org/0000-0002-0476-8020"},"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":"Tie-Yan Liu","raw_affiliation_strings":["Microsoft Research Asia"],"affiliations":[{"raw_affiliation_string":"Microsoft Research Asia","institution_ids":["https://openalex.org/I4210113369"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":8,"corresponding_author_ids":["https://openalex.org/A5030131487"],"corresponding_institution_ids":["https://openalex.org/I126520041"],"apc_list":null,"apc_paid":null,"fwci":6.8997,"has_fulltext":false,"cited_by_count":62,"citation_normalized_percentile":{"value":0.97367113,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":89,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"5699","last_page":"5703"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9998999834060669,"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"}},"topics":[{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9998999834060669,"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":0.9993000030517578,"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":0.9962000250816345,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8747294545173645},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6925753355026245},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.6185423731803894},{"id":"https://openalex.org/keywords/autoregressive-model","display_name":"Autoregressive model","score":0.5756320357322693},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.5466287136077881},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.4834331274032593},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.452680766582489},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.4257071316242218},{"id":"https://openalex.org/keywords/architecture","display_name":"Architecture","score":0.4102444052696228},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3148496448993683},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.18025904893875122},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.09524008631706238}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8747294545173645},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6925753355026245},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.6185423731803894},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.5756320357322693},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.5466287136077881},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.4834331274032593},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.452680766582489},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4257071316242218},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.4102444052696228},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3148496448993683},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.18025904893875122},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.09524008631706238},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","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},{"id":"https://openalex.org/C153349607","wikidata":"https://www.wikidata.org/wiki/Q36649","display_name":"Visual arts","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C142362112","wikidata":"https://www.wikidata.org/wiki/Q735","display_name":"Art","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp39728.2021.9414403","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9414403","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4","score":0.5299999713897705}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":44,"referenced_works":["https://openalex.org/W278779762","https://openalex.org/W349236604","https://openalex.org/W1902934009","https://openalex.org/W2469230926","https://openalex.org/W2625477277","https://openalex.org/W2885820039","https://openalex.org/W2888429796","https://openalex.org/W2903739847","https://openalex.org/W2945613576","https://openalex.org/W2946200149","https://openalex.org/W2953333557","https://openalex.org/W2955425717","https://openalex.org/W2962746461","https://openalex.org/W2963114950","https://openalex.org/W2963363373","https://openalex.org/W2963609956","https://openalex.org/W2963691546","https://openalex.org/W2964081807","https://openalex.org/W2964243274","https://openalex.org/W2970730223","https://openalex.org/W3000103182","https://openalex.org/W3007549584","https://openalex.org/W3015282541","https://openalex.org/W3015338123","https://openalex.org/W3025793647","https://openalex.org/W3033411150","https://openalex.org/W3034560159","https://openalex.org/W3041574813","https://openalex.org/W3094801149","https://openalex.org/W3096442195","https://openalex.org/W4300687381","https://openalex.org/W6639703010","https://openalex.org/W6748057086","https://openalex.org/W6749489859","https://openalex.org/W6753344092","https://openalex.org/W6753701838","https://openalex.org/W6762718338","https://openalex.org/W6763832098","https://openalex.org/W6765069029","https://openalex.org/W6770616408","https://openalex.org/W6774067568","https://openalex.org/W6778823374","https://openalex.org/W6780658294","https://openalex.org/W6917585676"],"related_works":["https://openalex.org/W2806676437","https://openalex.org/W2996843693","https://openalex.org/W2969565141","https://openalex.org/W4320854193","https://openalex.org/W3082358554","https://openalex.org/W3131663603","https://openalex.org/W2981519481","https://openalex.org/W3093101573","https://openalex.org/W2999012726","https://openalex.org/W4372263604"],"abstract_inverted_index":{"Text":[0],"to":[1,8,82,120],"speech":[2,13],"(TTS)":[3],"has":[4],"been":[5],"broadly":[6],"used":[7],"synthesize":[9],"natural":[10],"and":[11,35,57,87,102,112,144],"intelligible":[12],"in":[14,19,66],"different":[15],"scenarios.":[16],"Deploying":[17],"TTS":[18,40],"various":[20,110],"end":[21],"devices":[22,29],"such":[23,42],"as":[24,43],"mobile":[25],"phones":[26],"or":[27],"embedded":[28],"requires":[30],"extremely":[31],"small":[32],"memory":[33],"usage":[34],"inference":[36,49,58,146],"latency.":[37],"While":[38],"non-autoregressive":[39],"models":[41,89],"FastSpeech":[44],"have":[45],"achieved":[46],"significantly":[47],"faster":[48],"speed":[50],"than":[51],"autoregressive":[52],"models,":[53],"their":[54],"model":[55,101,134,141],"size":[56],"latency":[59],"are":[60,157],"still":[61],"large":[62],"for":[63],"the":[64,96,127,133],"deployment":[65],"resource":[67],"constrained":[68],"devices.":[69],"In":[70],"this":[71],"paper,":[72],"we":[73],"propose":[74],"LightSpeech,":[75],"which":[76],"leverages":[77],"neural":[78],"architecture":[79],"search":[80,107,128],"(NAS)":[81],"automatically":[83,121],"design":[84,104],"more":[85],"lightweight":[86,111],"efficient":[88],"based":[90],"on":[91,148,151],"FastSpeech.":[92],"We":[93],"first":[94],"profile":[95],"components":[97],"of":[98],"current":[99],"Fast-Speech":[100],"carefully":[103],"a":[105],"novel":[106],"space":[108],"containing":[109],"potentially":[113],"effective":[114],"architectures.":[115],"Then":[116],"NAS":[117],"is":[118],"utilized":[119],"discover":[122],"well":[123],"performing":[124],"architectures":[125],"within":[126],"space.":[129],"Experiments":[130],"show":[131],"that":[132],"discovered":[135],"by":[136],"our":[137],"method":[138],"achieves":[139],"15x":[140],"compression":[142],"ratio":[143],"6.5x":[145],"speedup":[147],"CPU":[149],"with":[150],"par":[152],"voice":[153],"quality.":[154],"Audio":[155],"demos":[156],"provided":[158],"at":[159],"https://speechresearch.github.io/lightspeech.":[160]},"counts_by_year":[{"year":2025,"cited_by_count":12},{"year":2024,"cited_by_count":9},{"year":2023,"cited_by_count":23},{"year":2022,"cited_by_count":12},{"year":2021,"cited_by_count":5},{"year":2020,"cited_by_count":1}],"updated_date":"2026-04-03T22:45:19.894376","created_date":"2025-10-10T00:00:00"}