{"id":"https://openalex.org/W7148283556","doi":"https://doi.org/10.1109/asru65441.2025.11434732","title":"Text-Guided Speech Representations for Language Acquisition Assessment","display_name":"Text-Guided Speech Representations for Language Acquisition Assessment","publication_year":2025,"publication_date":"2025-12-06","ids":{"openalex":"https://openalex.org/W7148283556","doi":"https://doi.org/10.1109/asru65441.2025.11434732"},"language":null,"primary_location":{"id":"doi:10.1109/asru65441.2025.11434732","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asru65441.2025.11434732","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","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/A5132741265","display_name":"Ilja Baumann","orcid":null},"institutions":[{"id":"https://openalex.org/I174004417","display_name":"Munich University of Applied Sciences","ror":"https://ror.org/012k1v959","country_code":"DE","type":"education","lineage":["https://openalex.org/I174004417"]}],"countries":["DE"],"is_corresponding":true,"raw_author_name":"Ilja Baumann","raw_affiliation_strings":["Technische Hochschule,Nürnberg"],"affiliations":[{"raw_affiliation_string":"Technische Hochschule,Nürnberg","institution_ids":["https://openalex.org/I174004417"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5132745922","display_name":"Dominik Wagner","orcid":null},"institutions":[{"id":"https://openalex.org/I174004417","display_name":"Munich University of Applied Sciences","ror":"https://ror.org/012k1v959","country_code":"DE","type":"education","lineage":["https://openalex.org/I174004417"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Dominik Wagner","raw_affiliation_strings":["Technische Hochschule,Nürnberg"],"affiliations":[{"raw_affiliation_string":"Technische Hochschule,Nürnberg","institution_ids":["https://openalex.org/I174004417"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5062589730","display_name":"Philipp Seeberger","orcid":"https://orcid.org/0009-0003-3161-5120"},"institutions":[{"id":"https://openalex.org/I174004417","display_name":"Munich University of Applied Sciences","ror":"https://ror.org/012k1v959","country_code":"DE","type":"education","lineage":["https://openalex.org/I174004417"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Philipp Seeberger","raw_affiliation_strings":["Technische Hochschule,Nürnberg"],"affiliations":[{"raw_affiliation_string":"Technische Hochschule,Nürnberg","institution_ids":["https://openalex.org/I174004417"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5053498662","display_name":"Korbinian Riedhammer","orcid":"https://orcid.org/0000-0003-3582-2154"},"institutions":[{"id":"https://openalex.org/I174004417","display_name":"Munich University of Applied Sciences","ror":"https://ror.org/012k1v959","country_code":"DE","type":"education","lineage":["https://openalex.org/I174004417"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Korbinian Riedhammer","raw_affiliation_strings":["Technische Hochschule,Nürnberg"],"affiliations":[{"raw_affiliation_string":"Technische Hochschule,Nürnberg","institution_ids":["https://openalex.org/I174004417"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5033302750","display_name":"Tobias Bocklet","orcid":"https://orcid.org/0009-0008-7780-8821"},"institutions":[{"id":"https://openalex.org/I174004417","display_name":"Munich University of Applied Sciences","ror":"https://ror.org/012k1v959","country_code":"DE","type":"education","lineage":["https://openalex.org/I174004417"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Tobias Bocklet","raw_affiliation_strings":["Technische Hochschule,Nürnberg"],"affiliations":[{"raw_affiliation_string":"Technische Hochschule,Nürnberg","institution_ids":["https://openalex.org/I174004417"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5132741265"],"corresponding_institution_ids":["https://openalex.org/I174004417"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.87549489,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"8"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.5817000269889832,"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":0.5817000269889832,"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/T10667","display_name":"Emotion and Mood Recognition","score":0.0568000003695488,"subfield":{"id":"https://openalex.org/subfields/3205","display_name":"Experimental and Cognitive Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T12031","display_name":"Speech and dialogue systems","score":0.04280000180006027,"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/natural-language","display_name":"Natural language","score":0.36820000410079956},{"id":"https://openalex.org/keywords/language-acquisition","display_name":"Language acquisition","score":0.3522999882698059},{"id":"https://openalex.org/keywords/identification","display_name":"Identification (biology)","score":0.27149999141693115},{"id":"https://openalex.org/keywords/language-identification","display_name":"Language identification","score":0.2558000087738037}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5776000022888184},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.439300000667572},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.3750999867916107},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.3734000027179718},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.36820000410079956},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3582000136375427},{"id":"https://openalex.org/C74672266","wikidata":"https://www.wikidata.org/wiki/Q815859","display_name":"Language acquisition","level":2,"score":0.3522999882698059},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.3253999948501587},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.27149999141693115},{"id":"https://openalex.org/C129792486","wikidata":"https://www.wikidata.org/wiki/Q1050419","display_name":"Language identification","level":3,"score":0.2558000087738037},{"id":"https://openalex.org/C163985040","wikidata":"https://www.wikidata.org/wiki/Q1172399","display_name":"Data acquisition","level":2,"score":0.2556000053882599}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/asru65441.2025.11434732","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asru65441.2025.11434732","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education","score":0.6164327263832092}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":28,"referenced_works":["https://openalex.org/W293305725","https://openalex.org/W2139008940","https://openalex.org/W2283896980","https://openalex.org/W2316843892","https://openalex.org/W2696967604","https://openalex.org/W2768358742","https://openalex.org/W2963425185","https://openalex.org/W2970641574","https://openalex.org/W3037671277","https://openalex.org/W3100806282","https://openalex.org/W3205644108","https://openalex.org/W3209059054","https://openalex.org/W3209984917","https://openalex.org/W4200300291","https://openalex.org/W4226033575","https://openalex.org/W4286359908","https://openalex.org/W4296069350","https://openalex.org/W4321021159","https://openalex.org/W4372260307","https://openalex.org/W4385245566","https://openalex.org/W4385987229","https://openalex.org/W4385987567","https://openalex.org/W4391021651","https://openalex.org/W4392903148","https://openalex.org/W4402111252","https://openalex.org/W4402112143","https://openalex.org/W4402112533","https://openalex.org/W4402116601"],"related_works":[],"abstract_inverted_index":{"Assessing":[0],"spoken":[1,64],"language":[2,29,126],"abilities":[3],"in":[4,27],"children":[5],"is":[6],"critical":[7],"for":[8,19,101,120],"early":[9,125],"detection":[10],"of":[11,25,108,124],"developmental":[12],"delays.":[13],"We":[14],"propose":[15],"a":[16,28,38,48,97],"two-stage":[17],"framework":[18],"automatic":[20],"pronunciation":[21],"assessment,":[22],"targeting":[23],"classification":[24,93],"items":[26],"development":[30],"screening.":[31],"In":[32,53],"the":[33,54,67,91,106,118],"first":[34],"stage,":[35,56],"we":[36,57],"fine-tune":[37,58],"sentence":[39],"transformer":[40],"using":[41],"phonetic":[42],"and":[43,95],"label-informed":[44],"losses":[45],"to":[46,62,85],"create":[47],"structured":[49],"latent":[50,98],"text":[51,68],"space.":[52,69],"second":[55],"an":[59],"audio":[60,75],"encoder":[61],"project":[63],"utterances":[65],"into":[66],"Our":[70],"approach":[71],"enables":[72],"inference":[73],"from":[74,114],"alone":[76],"while":[77],"preserving":[78],"alignment":[79],"with":[80],"phonetically":[81],"meaningful":[82],"structure.":[83],"Compared":[84],"baseline":[86],"systems,":[87],"our":[88],"method":[89],"achieves":[90],"best":[92],"performance":[94],"provides":[96],"space":[99],"suitable":[100],"downstream":[102],"analysis.":[103],"This":[104],"facilitates":[105],"identification":[107],"phonological":[109],"or":[110],"grammatical":[111],"error":[112],"patterns":[113],"cluster":[115],"structure,":[116],"paving":[117],"way":[119],"potentially":[121],"interpretable":[122],"assessment":[123],"acquisition.":[127]},"counts_by_year":[],"updated_date":"2026-04-03T16:44:17.987007","created_date":"2026-04-03T00:00:00"}