{"id":"https://openalex.org/W4394585787","doi":"https://doi.org/10.1109/lsp.2024.3386492","title":"Instabilities in Convnets for Raw Audio","display_name":"Instabilities in Convnets for Raw Audio","publication_year":2024,"publication_date":"2024-01-01","ids":{"openalex":"https://openalex.org/W4394585787","doi":"https://doi.org/10.1109/lsp.2024.3386492"},"language":"en","primary_location":{"id":"doi:10.1109/lsp.2024.3386492","is_oa":false,"landing_page_url":"https://doi.org/10.1109/lsp.2024.3386492","pdf_url":null,"source":{"id":"https://openalex.org/S120629676","display_name":"IEEE Signal Processing Letters","issn_l":"1070-9908","issn":["1070-9908","1558-2361"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Signal Processing Letters","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://hal.science/hal-04528116v1/document","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5077204791","display_name":"Daniel Haider","orcid":"https://orcid.org/0000-0001-8012-5521"},"institutions":[{"id":"https://openalex.org/I138211613","display_name":"Austrian Academy of Sciences","ror":"https://ror.org/03anc3s24","country_code":"AT","type":"government","lineage":["https://openalex.org/I138211613"]},{"id":"https://openalex.org/I2800119247","display_name":"Acoustics Research Institute","ror":"https://ror.org/04jd9ff79","country_code":"AT","type":"facility","lineage":["https://openalex.org/I138211613","https://openalex.org/I2800119247"]}],"countries":["AT"],"is_corresponding":true,"raw_author_name":"Daniel Haider","raw_affiliation_strings":["Acoustics Research Institute, Austrian Academy of Sciences, Vienna, Austria"],"affiliations":[{"raw_affiliation_string":"Acoustics Research Institute, Austrian Academy of Sciences, Vienna, Austria","institution_ids":["https://openalex.org/I138211613","https://openalex.org/I2800119247"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027789239","display_name":"Vincent Lostanlen","orcid":"https://orcid.org/0000-0003-0580-1651"},"institutions":[{"id":"https://openalex.org/I100445878","display_name":"\u00c9cole Centrale de Nantes","ror":"https://ror.org/03nh7d505","country_code":"FR","type":"education","lineage":["https://openalex.org/I100445878","https://openalex.org/I97188460"]},{"id":"https://openalex.org/I1294671590","display_name":"Centre National de la Recherche Scientifique","ror":"https://ror.org/02feahw73","country_code":"FR","type":"government","lineage":["https://openalex.org/I1294671590"]},{"id":"https://openalex.org/I4210117005","display_name":"Laboratoire des Sciences du Num\u00e9rique de Nantes","ror":"https://ror.org/02snf8m58","country_code":"FR","type":"facility","lineage":["https://openalex.org/I100445878","https://openalex.org/I1294671590","https://openalex.org/I1294671590","https://openalex.org/I1294671590","https://openalex.org/I1326498283","https://openalex.org/I205703379","https://openalex.org/I4210117005","https://openalex.org/I4210124215","https://openalex.org/I4210127572","https://openalex.org/I4210139971","https://openalex.org/I97188460","https://openalex.org/I97188460"]},{"id":"https://openalex.org/I97188460","display_name":"Nantes Universit\u00e9","ror":"https://ror.org/03gnr7b55","country_code":"FR","type":"education","lineage":["https://openalex.org/I97188460"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Vincent Lostanlen","raw_affiliation_strings":["CNRS, LS2N, UMR 6004, &#x00C9;cole Centrale Nantes, Nantes Universit&#x00E9;, Nantes, France"],"affiliations":[{"raw_affiliation_string":"CNRS, LS2N, UMR 6004, &#x00C9;cole Centrale Nantes, Nantes Universit&#x00E9;, Nantes, France","institution_ids":["https://openalex.org/I97188460","https://openalex.org/I100445878","https://openalex.org/I4210117005","https://openalex.org/I1294671590"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029839615","display_name":"Martin Ehler","orcid":"https://orcid.org/0000-0002-3247-6279"},"institutions":[{"id":"https://openalex.org/I129774422","display_name":"University of Vienna","ror":"https://ror.org/03prydq77","country_code":"AT","type":"education","lineage":["https://openalex.org/I129774422"]}],"countries":["AT"],"is_corresponding":false,"raw_author_name":"Martin Ehler","raw_affiliation_strings":["Faculty of Mathematics, University of Vienna, Vienna, Austria"],"affiliations":[{"raw_affiliation_string":"Faculty of Mathematics, University of Vienna, Vienna, Austria","institution_ids":["https://openalex.org/I129774422"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5084861530","display_name":"P\u00e9ter Bal\u00e1zs","orcid":"https://orcid.org/0000-0003-4939-0831"},"institutions":[{"id":"https://openalex.org/I138211613","display_name":"Austrian Academy of Sciences","ror":"https://ror.org/03anc3s24","country_code":"AT","type":"government","lineage":["https://openalex.org/I138211613"]},{"id":"https://openalex.org/I2800119247","display_name":"Acoustics Research Institute","ror":"https://ror.org/04jd9ff79","country_code":"AT","type":"facility","lineage":["https://openalex.org/I138211613","https://openalex.org/I2800119247"]}],"countries":["AT"],"is_corresponding":false,"raw_author_name":"Peter Balazs","raw_affiliation_strings":["Acoustics Research Institute, Austrian Academy of Sciences, Vienna, Austria"],"affiliations":[{"raw_affiliation_string":"Acoustics Research Institute, Austrian Academy of Sciences, Vienna, Austria","institution_ids":["https://openalex.org/I138211613","https://openalex.org/I2800119247"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5077204791"],"corresponding_institution_ids":["https://openalex.org/I138211613","https://openalex.org/I2800119247"],"apc_list":null,"apc_paid":null,"fwci":0.7257,"has_fulltext":true,"cited_by_count":2,"citation_normalized_percentile":{"value":0.65334813,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":96},"biblio":{"volume":"31","issue":null,"first_page":"1084","last_page":"1088"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":0.9980999827384949,"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"}},"topics":[{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":0.9980999827384949,"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/T11349","display_name":"Music Technology and Sound Studies","score":0.9980000257492065,"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/T11309","display_name":"Music and Audio Processing","score":0.9979000091552734,"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":0.6252975463867188},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.38210228085517883},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3460700213909149}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6252975463867188},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.38210228085517883},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3460700213909149}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/lsp.2024.3386492","is_oa":false,"landing_page_url":"https://doi.org/10.1109/lsp.2024.3386492","pdf_url":null,"source":{"id":"https://openalex.org/S120629676","display_name":"IEEE Signal Processing Letters","issn_l":"1070-9908","issn":["1070-9908","1558-2361"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Signal Processing Letters","raw_type":"journal-article"},{"id":"pmh:oai:HAL:hal-04528116v1","is_oa":true,"landing_page_url":"https://hal.science/hal-04528116","pdf_url":"https://hal.science/hal-04528116v1/document","source":{"id":"https://openalex.org/S4406922454","display_name":"SPIRE - Sciences Po Institutional REpository","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"IEEE Signal Processing Letters, 2024, 31, pp.1084-1088. &#x27E8;10.1109/LSP.2024.3386492&#x27E9;","raw_type":"Journal articles"}],"best_oa_location":{"id":"pmh:oai:HAL:hal-04528116v1","is_oa":true,"landing_page_url":"https://hal.science/hal-04528116","pdf_url":"https://hal.science/hal-04528116v1/document","source":{"id":"https://openalex.org/S4406922454","display_name":"SPIRE - Sciences Po Institutional REpository","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"IEEE Signal Processing Letters, 2024, 31, pp.1084-1088. &#x27E8;10.1109/LSP.2024.3386492&#x27E9;","raw_type":"Journal articles"},"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320320883","display_name":"Agence Nationale de la Recherche","ror":"https://ror.org/00rbzpz17"},{"id":"https://openalex.org/F4320321003","display_name":"Vienna Science and Technology Fund","ror":"https://ror.org/01f9mc681"},{"id":"https://openalex.org/F4320321004","display_name":"\u00d6sterreichischen Akademie der Wissenschaften","ror":"https://ror.org/03anc3s24"},{"id":"https://openalex.org/F4320321181","display_name":"Austrian Science Fund","ror":"https://ror.org/013tf3c58"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4394585787.pdf","grobid_xml":"https://content.openalex.org/works/W4394585787.grobid-xml"},"referenced_works_count":32,"referenced_works":["https://openalex.org/W228380312","https://openalex.org/W563126609","https://openalex.org/W1560153690","https://openalex.org/W1607202247","https://openalex.org/W1761550676","https://openalex.org/W1914942147","https://openalex.org/W2042587503","https://openalex.org/W2058770478","https://openalex.org/W2086874647","https://openalex.org/W2109246257","https://openalex.org/W2132360759","https://openalex.org/W2398826216","https://openalex.org/W2547304569","https://openalex.org/W2581711117","https://openalex.org/W2612464443","https://openalex.org/W2753908209","https://openalex.org/W2784225997","https://openalex.org/W2891171253","https://openalex.org/W2962901777","https://openalex.org/W2964052309","https://openalex.org/W3114848746","https://openalex.org/W3190935952","https://openalex.org/W4214655602","https://openalex.org/W4223504836","https://openalex.org/W4226051760","https://openalex.org/W4283803331","https://openalex.org/W4285198546","https://openalex.org/W4312856903","https://openalex.org/W4386614006","https://openalex.org/W6736987314","https://openalex.org/W6765807149","https://openalex.org/W6789403026"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052","https://openalex.org/W2382290278","https://openalex.org/W4395014643"],"abstract_inverted_index":{"What":[0],"makes":[1],"waveform-based":[2],"deep":[3],"learning":[4],"so":[5],"hard?":[6],"Despite":[7],"numerous":[8],"attempts":[9],"at":[10],"training":[11],"convolutional":[12,119],"neural":[13],"networks":[14],"(convnets)":[15],"for":[16,72,88,100],"filterbank":[17],"design,":[18],"they":[19,34],"often":[20],"fail":[21],"to":[22,50],"outperform":[23],"hand-crafted":[24],"baselines.":[25],"These":[26],"baselines":[27],"are":[28,97],"linear":[29],"time-invariant":[30],"systems:":[31],"as":[32],"such,":[33],"can":[35],"be":[36],"approximated":[37],"by":[38],"convnets":[39],"with":[40,79,108],"wide":[41],"receptive":[42],"fields.":[43],"Yet,":[44],"in":[45],"practice,":[46],"gradient-based":[47],"optimization":[48],"leads":[49],"suboptimal":[51],"results.":[52],"In":[53],"our":[54,109],"article,":[55],"we":[56],"approach":[57],"this":[58],"problem":[59],"from":[60],"the":[61,73,114,127,132],"perspective":[62],"of":[63,69,76,117,131,137],"initialization.":[64],"We":[65,83],"present":[66],"a":[67,118,122],"theory":[68,110],"large":[70,89],"deviations":[71,86],"energy":[74],"response":[75],"FIR":[77],"filterbanks":[78],"random":[80],"Gaussian":[81],"weights.":[82],"find":[84],"that":[85,113],"worsen":[87],"filters":[90],"and":[91,111,129],"locally":[92],"periodic":[93],"input":[94],"signals,":[95],"which":[96,134],"both":[98],"typical":[99],"audio":[101],"signal":[102],"processing":[103],"applications.":[104],"Numerical":[105],"simulations":[106],"align":[107],"suggest":[112],"condition":[115],"number":[116,128],"layer":[120],"follows":[121],"logarithmic":[123],"scaling":[124],"law":[125],"between":[126],"length":[130],"filters,":[133],"is":[135],"reminiscent":[136],"discrete":[138],"wavelet":[139],"bases.":[140]},"counts_by_year":[{"year":2025,"cited_by_count":2}],"updated_date":"2026-03-13T16:22:10.518609","created_date":"2025-10-10T00:00:00"}
