{"id":"https://openalex.org/W4372260183","doi":"https://doi.org/10.1109/icassp49357.2023.10096566","title":"StreamSpeech: Low-Latency Neural Architecture for High-Quality on-Device Speech Synthesis","display_name":"StreamSpeech: Low-Latency Neural Architecture for High-Quality on-Device Speech Synthesis","publication_year":2023,"publication_date":"2023-05-05","ids":{"openalex":"https://openalex.org/W4372260183","doi":"https://doi.org/10.1109/icassp49357.2023.10096566"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49357.2023.10096566","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49357.2023.10096566","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2023 - 2023 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/A5008307349","display_name":"Georgi Shopov","orcid":"https://orcid.org/0000-0002-7649-0064"},"institutions":[{"id":"https://openalex.org/I24768866","display_name":"Bulgarian Academy of Sciences","ror":"https://ror.org/01x8hew03","country_code":"BG","type":"funder","lineage":["https://openalex.org/I24768866"]},{"id":"https://openalex.org/I4210163063","display_name":"Institute of Information and Communication Technologies","ror":"https://ror.org/05fpsjc82","country_code":"BG","type":"facility","lineage":["https://openalex.org/I24768866","https://openalex.org/I4210163063"]}],"countries":["BG"],"is_corresponding":true,"raw_author_name":"Georgi Shopov","raw_affiliation_strings":["Institute of Information and Communication Technologies,Bulgarian Academy of Sciences","Bulgarian Academy of Sciences, Institute of Information and Communication Technologies"],"affiliations":[{"raw_affiliation_string":"Institute of Information and Communication Technologies,Bulgarian Academy of Sciences","institution_ids":["https://openalex.org/I4210163063"]},{"raw_affiliation_string":"Bulgarian Academy of Sciences, Institute of Information and Communication Technologies","institution_ids":["https://openalex.org/I4210163063","https://openalex.org/I24768866"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5006721962","display_name":"Stefan Gerdjikov","orcid":"https://orcid.org/0000-0002-5857-0466"},"institutions":[{"id":"https://openalex.org/I4210163063","display_name":"Institute of Information and Communication Technologies","ror":"https://ror.org/05fpsjc82","country_code":"BG","type":"facility","lineage":["https://openalex.org/I24768866","https://openalex.org/I4210163063"]},{"id":"https://openalex.org/I24768866","display_name":"Bulgarian Academy of Sciences","ror":"https://ror.org/01x8hew03","country_code":"BG","type":"funder","lineage":["https://openalex.org/I24768866"]},{"id":"https://openalex.org/I58918642","display_name":"Sofia University \"St. Kliment Ohridski\"","ror":"https://ror.org/02jv3k292","country_code":"BG","type":"education","lineage":["https://openalex.org/I58918642"]}],"countries":["BG"],"is_corresponding":false,"raw_author_name":"Stefan Gerdjikov","raw_affiliation_strings":["Institute of Information and Communication Technologies,Bulgarian Academy of Sciences","Bulgarian Academy of Sciences, Institute of Information and Communication Technologies","Faculty of Mathematics and Informatics, Sofia University, Bulgaria"],"affiliations":[{"raw_affiliation_string":"Institute of Information and Communication Technologies,Bulgarian Academy of Sciences","institution_ids":["https://openalex.org/I4210163063"]},{"raw_affiliation_string":"Bulgarian Academy of Sciences, Institute of Information and Communication Technologies","institution_ids":["https://openalex.org/I4210163063","https://openalex.org/I24768866"]},{"raw_affiliation_string":"Faculty of Mathematics and Informatics, Sofia University, Bulgaria","institution_ids":["https://openalex.org/I58918642"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5114255303","display_name":"Stoyan Mihov","orcid":null},"institutions":[{"id":"https://openalex.org/I24768866","display_name":"Bulgarian Academy of Sciences","ror":"https://ror.org/01x8hew03","country_code":"BG","type":"funder","lineage":["https://openalex.org/I24768866"]},{"id":"https://openalex.org/I4210163063","display_name":"Institute of Information and Communication Technologies","ror":"https://ror.org/05fpsjc82","country_code":"BG","type":"facility","lineage":["https://openalex.org/I24768866","https://openalex.org/I4210163063"]}],"countries":["BG"],"is_corresponding":false,"raw_author_name":"Stoyan Mihov","raw_affiliation_strings":["Institute of Information and Communication Technologies,Bulgarian Academy of Sciences","Bulgarian Academy of Sciences, Institute of Information and Communication Technologies"],"affiliations":[{"raw_affiliation_string":"Institute of Information and Communication Technologies,Bulgarian Academy of Sciences","institution_ids":["https://openalex.org/I4210163063"]},{"raw_affiliation_string":"Bulgarian Academy of Sciences, Institute of Information and Communication Technologies","institution_ids":["https://openalex.org/I4210163063","https://openalex.org/I24768866"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5008307349"],"corresponding_institution_ids":["https://openalex.org/I24768866","https://openalex.org/I4210163063"],"apc_list":null,"apc_paid":null,"fwci":0.1748,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.52326631,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":95},"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/T10201","display_name":"Speech Recognition and Synthesis","score":1.0,"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":1.0,"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":0.9976999759674072,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9976000189781189,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8146342039108276},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.7737325429916382},{"id":"https://openalex.org/keywords/x86","display_name":"x86","score":0.6397830843925476},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.4636353850364685},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.4437633156776428},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.3792383074760437},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.33809036016464233},{"id":"https://openalex.org/keywords/real-time-computing","display_name":"Real-time computing","score":0.32480549812316895},{"id":"https://openalex.org/keywords/software","display_name":"Software","score":0.2084406316280365},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.16837313771247864},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.11394822597503662}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8146342039108276},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.7737325429916382},{"id":"https://openalex.org/C170723468","wikidata":"https://www.wikidata.org/wiki/Q182933","display_name":"x86","level":3,"score":0.6397830843925476},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.4636353850364685},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.4437633156776428},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.3792383074760437},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.33809036016464233},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.32480549812316895},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.2084406316280365},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.16837313771247864},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.11394822597503662},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49357.2023.10096566","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49357.2023.10096566","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2023 - 2023 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":0.4699999988079071}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":28,"referenced_works":["https://openalex.org/W1966147466","https://openalex.org/W2121089405","https://openalex.org/W2806900363","https://openalex.org/W2953333557","https://openalex.org/W2963091184","https://openalex.org/W2964243274","https://openalex.org/W2971753973","https://openalex.org/W3033411150","https://openalex.org/W3034949308","https://openalex.org/W3095419383","https://openalex.org/W3095851005","https://openalex.org/W3095999419","https://openalex.org/W3096702751","https://openalex.org/W3097376233","https://openalex.org/W3097777922","https://openalex.org/W3150572638","https://openalex.org/W3160248152","https://openalex.org/W3160919572","https://openalex.org/W3198061281","https://openalex.org/W4210684870","https://openalex.org/W4225290925","https://openalex.org/W4252420661","https://openalex.org/W6677763620","https://openalex.org/W6765069029","https://openalex.org/W6767453231","https://openalex.org/W6778823374","https://openalex.org/W6784782250","https://openalex.org/W6785109475"],"related_works":["https://openalex.org/W3215381467","https://openalex.org/W4301207796","https://openalex.org/W2915956107","https://openalex.org/W2099986681","https://openalex.org/W4846490","https://openalex.org/W4240878335","https://openalex.org/W563303149","https://openalex.org/W2098290341","https://openalex.org/W4226057846","https://openalex.org/W2904058535"],"abstract_inverted_index":{"Neural":[0],"text-to-speech":[1],"(TTS)":[2],"systems":[3,24],"have":[4],"recently":[5],"demonstrated":[6],"the":[7,15,75,98,117,151,154],"ability":[8],"to":[9,85],"synthesize":[10],"high-quality":[11,53],"natural":[12],"speech.":[13,156],"However,":[14],"inference":[16],"latency":[17,63],"and":[18,55,89,110,134],"real-time":[19],"factor":[20],"(RTF)":[21],"of":[22,46,100,153],"such":[23],"are":[25],"still":[26],"too":[27],"high":[28],"for":[29],"deployment":[30],"on":[31,64,128,140],"devices":[32,66],"without":[33],"specialized":[34],"hardware.":[35],"In":[36],"this":[37],"paper,":[38],"we":[39],"describe":[40],"StreamSpeech":[41],"\u2013":[42],"an":[43],"optimized":[44],"architecture":[45],"a":[47,69,101,129,141],"complete":[48,119],"TTS":[49,77,120],"system":[50,121],"that":[51,116],"produces":[52],"speech":[54,112],"runs":[56],"faster":[57],"than":[58],"real":[59],"time":[60],"with":[61,83,146],"imperceptible":[62],"resource-constrained":[65],"by":[67],"utilizing":[68],"single":[70],"CPU":[71,133,145],"core.":[72],"We":[73],"divide":[74],"standard":[76],"processing":[78],"pipeline":[79],"into":[80],"three":[81],"phases":[82],"respect":[84],"their":[86],"operating":[87],"resolution":[88],"optimize":[90],"them":[91],"separately.":[92],"Our":[93],"main":[94],"novel":[95],"contribution":[96],"is":[97],"introduction":[99],"lightweight":[102],"convolutional":[103],"acoustic":[104],"model":[105],"decoder,":[106],"which":[107],"enables":[108],"streaming":[109],"low-latency":[111],"generation.":[113],"Experiments":[114],"show":[115],"resulting":[118],"achieves":[122],"79":[123],"ms":[124,136],"latency,":[125,137],"0.155":[126],"RTF":[127,139],"low-power":[130],"notebook":[131],"x86":[132],"276":[135],"0.289":[138],"mid-range":[142],"mobile":[143],"ARM":[144],"no":[147],"noticeable":[148],"difference":[149],"in":[150],"quality":[152],"generated":[155]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}
