{"id":"https://openalex.org/W4406496863","doi":"https://doi.org/10.1109/slt61566.2024.10832228","title":"Addressing Index Collapse of Large-Codebook Speech Tokenizer With Dual-Decoding Product-Quantized Variational Auto-Encoder","display_name":"Addressing Index Collapse of Large-Codebook Speech Tokenizer With Dual-Decoding Product-Quantized Variational Auto-Encoder","publication_year":2024,"publication_date":"2024-12-02","ids":{"openalex":"https://openalex.org/W4406496863","doi":"https://doi.org/10.1109/slt61566.2024.10832228"},"language":"en","primary_location":{"id":"doi:10.1109/slt61566.2024.10832228","is_oa":false,"landing_page_url":"https://doi.org/10.1109/slt61566.2024.10832228","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE Spoken Language Technology Workshop (SLT)","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/A5072447806","display_name":"Haohan Guo","orcid":"https://orcid.org/0000-0002-3393-9984"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"CN","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Haohan Guo","raw_affiliation_strings":["The Chinese University of Hong Kong,Hong Kong SAR,China"],"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong,Hong Kong SAR,China","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101692260","display_name":"Fenglong Xie","orcid":null},"institutions":[{"id":"https://openalex.org/I862669128","display_name":"Xiaomi (China)","ror":"https://ror.org/029f7bn57","country_code":"CN","type":"company","lineage":["https://openalex.org/I862669128"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Fenglong Xie","raw_affiliation_strings":["Xiaohongshu Inc,Shanghai,China"],"affiliations":[{"raw_affiliation_string":"Xiaohongshu Inc,Shanghai,China","institution_ids":["https://openalex.org/I862669128"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028178332","display_name":"Dongchao Yang","orcid":"https://orcid.org/0000-0003-0879-4047"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"CN","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Dongchao Yang","raw_affiliation_strings":["The Chinese University of Hong Kong,Hong Kong SAR,China"],"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong,Hong Kong SAR,China","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101486499","display_name":"Hui Lu","orcid":"https://orcid.org/0000-0002-7698-6262"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"CN","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hui Lu","raw_affiliation_strings":["The Chinese University of Hong Kong,Hong Kong SAR,China"],"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong,Hong Kong SAR,China","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5025119324","display_name":"Xixin Wu","orcid":"https://orcid.org/0000-0001-9543-1572"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"CN","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xixin Wu","raw_affiliation_strings":["The Chinese University of Hong Kong,Hong Kong SAR,China"],"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong,Hong Kong SAR,China","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5019458385","display_name":"Helen Meng","orcid":"https://orcid.org/0000-0002-4427-3532"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"CN","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Helen Meng","raw_affiliation_strings":["The Chinese University of Hong Kong,Hong Kong SAR,China"],"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong,Hong Kong SAR,China","institution_ids":["https://openalex.org/I177725633"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5072447806"],"corresponding_institution_ids":["https://openalex.org/I177725633"],"apc_list":null,"apc_paid":null,"fwci":1.0878,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.82775737,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":96},"biblio":{"volume":null,"issue":null,"first_page":"548","last_page":"553"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9994999766349792,"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":0.9994999766349792,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9972000122070312,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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":0.9952999949455261,"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/codebook","display_name":"Codebook","score":0.94573974609375},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.7520444393157959},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.6866001486778259},{"id":"https://openalex.org/keywords/dual","display_name":"Dual (grammatical number)","score":0.6650227308273315},{"id":"https://openalex.org/keywords/product","display_name":"Product (mathematics)","score":0.6130775213241577},{"id":"https://openalex.org/keywords/index","display_name":"Index (typography)","score":0.604629635810852},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5198567509651184},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.4275815784931183},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.398874431848526},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.33075907826423645},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.2663034200668335},{"id":"https://openalex.org/keywords/philosophy","display_name":"Philosophy","score":0.07992944121360779},{"id":"https://openalex.org/keywords/world-wide-web","display_name":"World Wide Web","score":0.06508761644363403},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.05847358703613281}],"concepts":[{"id":"https://openalex.org/C127759330","wikidata":"https://www.wikidata.org/wiki/Q637416","display_name":"Codebook","level":2,"score":0.94573974609375},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.7520444393157959},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.6866001486778259},{"id":"https://openalex.org/C2780980858","wikidata":"https://www.wikidata.org/wiki/Q110022","display_name":"Dual (grammatical number)","level":2,"score":0.6650227308273315},{"id":"https://openalex.org/C90673727","wikidata":"https://www.wikidata.org/wiki/Q901718","display_name":"Product (mathematics)","level":2,"score":0.6130775213241577},{"id":"https://openalex.org/C2777382242","wikidata":"https://www.wikidata.org/wiki/Q6017816","display_name":"Index (typography)","level":2,"score":0.604629635810852},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5198567509651184},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.4275815784931183},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.398874431848526},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.33075907826423645},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2663034200668335},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.07992944121360779},{"id":"https://openalex.org/C136764020","wikidata":"https://www.wikidata.org/wiki/Q466","display_name":"World Wide Web","level":1,"score":0.06508761644363403},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.05847358703613281},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/slt61566.2024.10832228","is_oa":false,"landing_page_url":"https://doi.org/10.1109/slt61566.2024.10832228","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE Spoken Language Technology Workshop (SLT)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320326427","display_name":"Innovation and Technology Fund","ror":null}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":35,"referenced_works":["https://openalex.org/W2119913432","https://openalex.org/W2124509324","https://openalex.org/W2752796333","https://openalex.org/W2935711438","https://openalex.org/W3001279689","https://openalex.org/W3097206152","https://openalex.org/W3197659778","https://openalex.org/W3203407300","https://openalex.org/W3209059054","https://openalex.org/W4226132755","https://openalex.org/W4313679638","https://openalex.org/W4367721746","https://openalex.org/W4376632512","https://openalex.org/W4379539302","https://openalex.org/W4386132131","https://openalex.org/W4386384714","https://openalex.org/W4387158294","https://openalex.org/W4390136555","https://openalex.org/W4390872598","https://openalex.org/W4391833199","https://openalex.org/W4399729833","https://openalex.org/W4402672068","https://openalex.org/W6772383348","https://openalex.org/W6785471559","https://openalex.org/W6838843145","https://openalex.org/W6848735303","https://openalex.org/W6852421699","https://openalex.org/W6852781825","https://openalex.org/W6852843773","https://openalex.org/W6852859116","https://openalex.org/W6855885476","https://openalex.org/W6856174626","https://openalex.org/W6856524194","https://openalex.org/W6857461762","https://openalex.org/W6859863615"],"related_works":["https://openalex.org/W2293149949","https://openalex.org/W2026099691","https://openalex.org/W4284672201","https://openalex.org/W2377486419","https://openalex.org/W2943202426","https://openalex.org/W2137816434","https://openalex.org/W2163679795","https://openalex.org/W2736714427","https://openalex.org/W2017956276","https://openalex.org/W2048606991"],"abstract_inverted_index":{"VQ-VAE,":[0],"as":[1],"a":[2,16,61,76],"mainstream":[3],"approach":[4],"of":[5,19],"speech":[6,45,49,129],"tokenizer,":[7],"has":[8],"been":[9],"troubled":[10],"by":[11],"\u201cindex":[12,93],"collapse\u201d,":[13],"where":[14],"only":[15],"small":[17],"number":[18],"codewords":[20,37,59],"are":[21],"activated":[22],"in":[23,60,124],"large":[24],"codebooks.":[25,99,133],"This":[26],"work":[27],"proposes":[28],"product-quantized":[29],"(PQ)":[30],"VAE":[31],"with":[32,80,102,131],"more":[33],"codebooks":[34],"but":[35],"fewer":[36],"to":[38,65],"address":[39],"this":[40],"problem":[41],"and":[42,55,83,111],"build":[43],"large-codebook":[44],"tokenizers.":[46],"It":[47],"encodes":[48],"features":[50],"into":[51,58],"multiple":[52],"VQ":[53,68,117],"subspaces":[54],"composes":[56],"them":[57],"larger":[62,98,132],"codebook.":[63],"Besides,":[64],"utilize":[66],"each":[67],"subspace":[69],"well,":[70],"we":[71],"also":[72],"enhance":[73],"PQ-VAE":[74,91,120],"via":[75],"dual-decoding":[77],"training":[78,105],"strategy":[79,106],"the":[81,103],"encoding":[82],"quantized":[84],"sequences.":[85],"The":[86,100],"experimental":[87],"results":[88],"demonstrate":[89],"that":[90],"addresses":[92],"collapse\u201d":[94],"effectively,":[95],"especially":[96],"for":[97],"model":[101],"proposed":[104],"further":[107],"improves":[108],"codebook":[109],"perplexity":[110],"reconstruction":[112],"quality,":[113],"outperforming":[114],"other":[115],"multi-codebook":[116],"approaches.":[118],"Finally,":[119],"demonstrates":[121],"its":[122],"effectiveness":[123],"language-model-based":[125],"TTS,":[126],"supporting":[127],"higher-quality":[128],"generation":[130]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":2}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}