{"id":"https://openalex.org/W4408356822","doi":"https://doi.org/10.1109/icassp49660.2025.10887992","title":"Uncertainty prediction for prominence classification with chroma features","display_name":"Uncertainty prediction for prominence classification with chroma features","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408356822","doi":"https://doi.org/10.1109/icassp49660.2025.10887992"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10887992","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10887992","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":"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/A5085707341","display_name":"Julian Linke","orcid":"https://orcid.org/0009-0002-9904-0944"},"institutions":[{"id":"https://openalex.org/I4092182","display_name":"Graz University of Technology","ror":"https://ror.org/00d7xrm67","country_code":"AT","type":"education","lineage":["https://openalex.org/I4092182"]}],"countries":["AT"],"is_corresponding":true,"raw_author_name":"Julian Linke","raw_affiliation_strings":["Graz University of Technology,Signal Processing and Speech Communication Laboratory,Austria"],"affiliations":[{"raw_affiliation_string":"Graz University of Technology,Signal Processing and Speech Communication Laboratory,Austria","institution_ids":["https://openalex.org/I4092182"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5023873185","display_name":"Sophie Steger","orcid":null},"institutions":[{"id":"https://openalex.org/I4092182","display_name":"Graz University of Technology","ror":"https://ror.org/00d7xrm67","country_code":"AT","type":"education","lineage":["https://openalex.org/I4092182"]}],"countries":["AT"],"is_corresponding":false,"raw_author_name":"Sophie Steger","raw_affiliation_strings":["Graz University of Technology,Signal Processing and Speech Communication Laboratory,Austria"],"affiliations":[{"raw_affiliation_string":"Graz University of Technology,Signal Processing and Speech Communication Laboratory,Austria","institution_ids":["https://openalex.org/I4092182"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5116594753","display_name":"Philipp Steinwender","orcid":null},"institutions":[{"id":"https://openalex.org/I4092182","display_name":"Graz University of Technology","ror":"https://ror.org/00d7xrm67","country_code":"AT","type":"education","lineage":["https://openalex.org/I4092182"]}],"countries":["AT"],"is_corresponding":false,"raw_author_name":"Philipp Steinwender","raw_affiliation_strings":["Graz University of Technology,Signal Processing and Speech Communication Laboratory,Austria"],"affiliations":[{"raw_affiliation_string":"Graz University of Technology,Signal Processing and Speech Communication Laboratory,Austria","institution_ids":["https://openalex.org/I4092182"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5071861893","display_name":"Gernot Kubin","orcid":"https://orcid.org/0000-0002-8420-6051"},"institutions":[{"id":"https://openalex.org/I4092182","display_name":"Graz University of Technology","ror":"https://ror.org/00d7xrm67","country_code":"AT","type":"education","lineage":["https://openalex.org/I4092182"]}],"countries":["AT"],"is_corresponding":false,"raw_author_name":"Gernot Kubin","raw_affiliation_strings":["Graz University of Technology,Signal Processing and Speech Communication Laboratory,Austria"],"affiliations":[{"raw_affiliation_string":"Graz University of Technology,Signal Processing and Speech Communication Laboratory,Austria","institution_ids":["https://openalex.org/I4092182"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015798259","display_name":"Franz Pernkopf","orcid":"https://orcid.org/0000-0002-6356-3367"},"institutions":[{"id":"https://openalex.org/I4092182","display_name":"Graz University of Technology","ror":"https://ror.org/00d7xrm67","country_code":"AT","type":"education","lineage":["https://openalex.org/I4092182"]}],"countries":["AT"],"is_corresponding":false,"raw_author_name":"Franz Pernkopf","raw_affiliation_strings":["Graz University of Technology,Signal Processing and Speech Communication Laboratory,Austria"],"affiliations":[{"raw_affiliation_string":"Graz University of Technology,Signal Processing and Speech Communication Laboratory,Austria","institution_ids":["https://openalex.org/I4092182"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5027851634","display_name":"Barbara Schuppler","orcid":"https://orcid.org/0000-0003-4009-0832"},"institutions":[{"id":"https://openalex.org/I4092182","display_name":"Graz University of Technology","ror":"https://ror.org/00d7xrm67","country_code":"AT","type":"education","lineage":["https://openalex.org/I4092182"]}],"countries":["AT"],"is_corresponding":false,"raw_author_name":"Barbara Schuppler","raw_affiliation_strings":["Graz University of Technology,Signal Processing and Speech Communication Laboratory,Austria"],"affiliations":[{"raw_affiliation_string":"Graz University of Technology,Signal Processing and Speech Communication Laboratory,Austria","institution_ids":["https://openalex.org/I4092182"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5085707341"],"corresponding_institution_ids":["https://openalex.org/I4092182"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.06582248,"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":"5"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12983","display_name":"Satellite Image Processing and Photogrammetry","score":0.739300012588501,"subfield":{"id":"https://openalex.org/subfields/2212","display_name":"Ocean Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T12983","display_name":"Satellite Image Processing and Photogrammetry","score":0.739300012588501,"subfield":{"id":"https://openalex.org/subfields/2212","display_name":"Ocean Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12389","display_name":"Infrared Target Detection Methodologies","score":0.7307999730110168,"subfield":{"id":"https://openalex.org/subfields/2202","display_name":"Aerospace Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5808638334274292},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5042022466659546},{"id":"https://openalex.org/keywords/statistical-classification","display_name":"Statistical classification","score":0.44393447041511536},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.42084968090057373},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.32589298486709595}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5808638334274292},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5042022466659546},{"id":"https://openalex.org/C110083411","wikidata":"https://www.wikidata.org/wiki/Q1744628","display_name":"Statistical classification","level":2,"score":0.44393447041511536},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.42084968090057373},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.32589298486709595}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49660.2025.10887992","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10887992","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":[{"id":"https://metadata.un.org/sdg/13","score":0.7900000214576721,"display_name":"Climate action"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320321181","display_name":"Austrian Science Fund","ror":"https://ror.org/013tf3c58"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":24,"referenced_works":["https://openalex.org/W1581480663","https://openalex.org/W1980993072","https://openalex.org/W1989674304","https://openalex.org/W2096429026","https://openalex.org/W2101234009","https://openalex.org/W2111959010","https://openalex.org/W2191779130","https://openalex.org/W2398044918","https://openalex.org/W2563546967","https://openalex.org/W2754499401","https://openalex.org/W2807799430","https://openalex.org/W2940820519","https://openalex.org/W3012395103","https://openalex.org/W3014596384","https://openalex.org/W3027574360","https://openalex.org/W4240805545","https://openalex.org/W4281752274","https://openalex.org/W6642382926","https://openalex.org/W6676368125","https://openalex.org/W6691503597","https://openalex.org/W6773104461","https://openalex.org/W6849594828","https://openalex.org/W6860962755","https://openalex.org/W7052236464"],"related_works":["https://openalex.org/W2961085424","https://openalex.org/W4306674287","https://openalex.org/W4387369504","https://openalex.org/W3046775127","https://openalex.org/W4394896187","https://openalex.org/W3170094116","https://openalex.org/W4386462264","https://openalex.org/W3107602296","https://openalex.org/W4364306694","https://openalex.org/W4312192474"],"abstract_inverted_index":{"This":[0,27],"paper":[1],"presents":[2],"methods":[3],"for":[4,32,94,113,122],"prominence":[5],"classification":[6,131],"in":[7,70],"conversational":[8,33],"speech.":[9,26],"Most":[10],"existing":[11],"tools":[12],"rely":[13],"on":[14,24,126],"prosodic":[15,63,95],"features":[16,64],"extracted":[17],"at":[18],"syllable-":[19],"or":[20],"phone-level,":[21],"performing":[22],"well":[23,56],"read":[25],"is":[28,41,120],"not":[29],"the":[30,36,71,74,86,92,100,117,123,134],"case":[31],"speech,":[34],"where":[35],"quality":[37],"of":[38,73,91],"automatic":[39],"segmentation":[40],"significantly":[42],"worse.":[43],"We":[44,78],"introduce":[45],"entropy-based":[46],"chroma":[47,97,127],"features,":[48,128],"requiring":[49],"only":[50],"word-level":[51],"segmentations.":[52],"They":[53],"perform":[54],"equally":[55],"as":[57,104],"a":[58],"random":[59],"forest":[60],"classifier":[61,124],"with":[62,68,107],"(requiring":[65],"phone-level":[66],"segmentation),":[67],"accuracies":[69],"range":[72],"human":[75],"inter-rater":[76,108],"agreement.":[77],"further":[79],"use":[80],"Bayesian":[81],"deep":[82],"learning":[83],"to":[84],"quantify":[85],"epistemic":[87,118],"and":[88,96,110],"aleatoric":[89,101],"uncertainty":[90,102,119],"prediction":[93],"features.":[98],"Whereas":[99],"is,":[103],"expected,":[105],"consistent":[106],"agreement":[109],"similarly":[111],"high":[112],"both":[114],"feature":[115],"sets,":[116],"lower":[121],"based":[125],"indicating":[129],"higher":[130],"consistency":[132],"across":[133],"corpus.":[135]},"counts_by_year":[],"updated_date":"2025-12-28T23:10:05.387466","created_date":"2025-10-10T00:00:00"}