{"id":"https://openalex.org/W4408354743","doi":"https://doi.org/10.1109/icassp49660.2025.10888963","title":"Lightweight neural front-ends for low-resource on-device Text-to-Speech","display_name":"Lightweight neural front-ends for low-resource on-device Text-to-Speech","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408354743","doi":"https://doi.org/10.1109/icassp49660.2025.10888963"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10888963","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10888963","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/A5041378836","display_name":"Giulia Comini","orcid":"https://orcid.org/0000-0002-9391-6565"},"institutions":[{"id":"https://openalex.org/I4210123934","display_name":"Amazon (United Kingdom)","ror":"https://ror.org/02xey9634","country_code":"GB","type":"company","lineage":["https://openalex.org/I1311688040","https://openalex.org/I4210123934"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Giulia Comini","raw_affiliation_strings":["Amazon AGI,United Kingdom"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Amazon AGI,United Kingdom","institution_ids":["https://openalex.org/I4210123934"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5053168102","display_name":"Heereen Shim","orcid":"https://orcid.org/0000-0003-3761-2310"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Heereen Shim","raw_affiliation_strings":["Amazon AGI,Netherlands"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Amazon AGI,Netherlands","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5109494794","display_name":"Manuel Sam Ribeiro","orcid":"https://orcid.org/0000-0001-8096-2231"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Manuel Sam Ribeiro","raw_affiliation_strings":["Amazon AGI,Poland"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Amazon AGI,Poland","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":null,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.9682999849319458,"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":0.9682999849319458,"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":0.9549999833106995,"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":0.6819754242897034},{"id":"https://openalex.org/keywords/front","display_name":"Front (military)","score":0.5579193830490112},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5110069513320923},{"id":"https://openalex.org/keywords/resource","display_name":"Resource (disambiguation)","score":0.49978184700012207},{"id":"https://openalex.org/keywords/front-and-back-ends","display_name":"Front and back ends","score":0.49729278683662415},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.1668281853199005},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.13635694980621338},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.09861454367637634},{"id":"https://openalex.org/keywords/mechanical-engineering","display_name":"Mechanical engineering","score":0.07282310724258423}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6819754242897034},{"id":"https://openalex.org/C2777551076","wikidata":"https://www.wikidata.org/wiki/Q842332","display_name":"Front (military)","level":2,"score":0.5579193830490112},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5110069513320923},{"id":"https://openalex.org/C206345919","wikidata":"https://www.wikidata.org/wiki/Q20380951","display_name":"Resource (disambiguation)","level":2,"score":0.49978184700012207},{"id":"https://openalex.org/C53016008","wikidata":"https://www.wikidata.org/wiki/Q620167","display_name":"Front and back ends","level":2,"score":0.49729278683662415},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.1668281853199005},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.13635694980621338},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.09861454367637634},{"id":"https://openalex.org/C78519656","wikidata":"https://www.wikidata.org/wiki/Q101333","display_name":"Mechanical engineering","level":1,"score":0.07282310724258423}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49660.2025.10888963","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10888963","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":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":34,"referenced_works":["https://openalex.org/W1618885279","https://openalex.org/W2103085228","https://openalex.org/W2164107060","https://openalex.org/W2317559641","https://openalex.org/W2962784628","https://openalex.org/W2963913768","https://openalex.org/W2972437137","https://openalex.org/W3037676302","https://openalex.org/W3095883095","https://openalex.org/W3154516348","https://openalex.org/W3160919572","https://openalex.org/W3197649190","https://openalex.org/W4221155853","https://openalex.org/W4224629880","https://openalex.org/W4312839555","https://openalex.org/W4385993937","https://openalex.org/W4389131500","https://openalex.org/W4391833199","https://openalex.org/W6638280674","https://openalex.org/W6679434410","https://openalex.org/W6684941089","https://openalex.org/W6729005282","https://openalex.org/W6747425110","https://openalex.org/W6766978945","https://openalex.org/W6769627184","https://openalex.org/W6795198151","https://openalex.org/W6800762109","https://openalex.org/W6843243982","https://openalex.org/W6852421699","https://openalex.org/W6854075300","https://openalex.org/W6855058360","https://openalex.org/W6855513303","https://openalex.org/W6857139653","https://openalex.org/W6860008077"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2074777875","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890"],"abstract_inverted_index":{"We":[0],"propose":[1],"a":[2,84,98],"lightweight":[3,71,114],"neural":[4,72],"front-end":[5,26],"framework":[6],"for":[7,58],"on-device":[8,59],"speech":[9],"generation":[10],"and":[11,41,56,74,125],"highlight":[12],"its":[13],"benefits":[14],"towards":[15],"low-resource":[16,118],"language":[17,34,119],"scaling.":[18],"While":[19],"data-driven":[20],"models":[21,76,111],"have":[22],"shown":[23],"potential":[24],"in":[25,53],"literature,":[27],"especially":[28],"since":[29],"they":[30,36],"can":[31],"enable":[32,117],"fast":[33],"expansion,":[35],"are":[37],"often":[38],"extremely":[39],"large":[40],"of":[42,65,88],"high":[43],"latency.":[44],"There":[45],"is":[46],"limited":[47],"work":[48],"focusing":[49],"on":[50,79,92,121],"their":[51],"usability":[52],"real-time":[54],"settings,":[55],"none":[57],"TTS":[60],"applications.":[61],"At":[62],"the":[63],"cost":[64],"small":[66],"performances":[67],"trade-offs,":[68],"we":[69,116],"build":[70],"Grapheme-to-Phoneme":[73,123],"verbalization":[75],"which":[77],"achieve,":[78],"average":[80],"across":[81],"three":[82],"languages,":[83],"p90":[85],"latency":[86],"reduction":[87],"95.98%":[89],"per":[90],"token":[91],"single-threaded":[93],"CPU@2.30GHz,":[94],"with":[95],"respect":[96],"to":[97,112],"traditional":[99],"transformer-based":[100],"baseline,":[101],"while":[102],"having":[103],"99.26%":[104],"less":[105],"parameters.":[106],"Additionally,":[107],"leveraging":[108],"pre-trained":[109],"teacher":[110],"bootstrap":[113],"students,":[115],"scaling":[120],"both":[122],"conversion":[124],"verbalization.":[126]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}