{"data":{"id":"10.6084/m9.figshare.1277821","type":"dois","attributes":{"doi":"10.6084/m9.figshare.1277821","prefix":"10.6084","suffix":"m9.figshare.1277821","identifiers":[],"alternateIdentifiers":[],"creators":[{"name":"Synnaeve, Gabriel","nameType":"Personal","givenName":"Gabriel","familyName":"Synnaeve","affiliation":[],"nameIdentifiers":[]},{"name":"Schatz, Thomas","nameType":"Personal","givenName":"Thomas","familyName":"Schatz","affiliation":[],"nameIdentifiers":[]},{"name":"Dupoux, Emmanuel","nameType":"Personal","givenName":"Emmanuel","familyName":"Dupoux","affiliation":[],"nameIdentifiers":[]}],"titles":[{"title":"Phonetics Embedding Learning with Side Information"}],"publisher":"figshare","container":{},"publicationYear":2014,"subjects":[{"subject":"Artificial Intelligence and Image Processing"},{"subject":"FOS: Computer and information sciences","schemeUri":"http://www.oecd.org/science/inno/38235147.pdf","subjectScheme":"Fields of Science and Technology (FOS)"},{"subject":"FOS: Computer and information sciences","subjectScheme":"Fields of Science and Technology (FOS)"}],"contributors":[],"dates":[{"date":"2014-12-23","dateType":"Created"},{"date":"2014-12-23","dateType":"Updated"},{"date":"2014","dateType":"Issued"}],"language":null,"types":{"ris":"DATA","bibtex":"misc","citeproc":"dataset","schemaOrg":"Dataset","resourceType":"Dataset","resourceTypeGeneral":"Dataset"},"relatedIdentifiers":[],"relatedItems":[],"sizes":["1310828 Bytes"],"formats":[],"version":null,"rightsList":[{"rights":"Creative Commons Attribution 4.0 International","rightsUri":"https://creativecommons.org/licenses/by/4.0/legalcode","schemeUri":"https://spdx.org/licenses/","rightsIdentifier":"cc-by-4.0","rightsIdentifierScheme":"SPDX"}],"descriptions":[{"description":"We show that it is possible to learn an efficient acoustic model using only a\u003cbr\u003esmall amount of easily available word-level similarity annotations. In contrast\u003cbr\u003eto the detailed phonetic labeling required by classical speech recognition\u003cbr\u003etechnologies, the only information our method requires are pairs of\u003cbr\u003espeech excerpts which are known to be similar (same word) and pairs of\u003cbr\u003espeech excerpts which are known to be different (different words). An acoustic model is obtained by training shallow and deep neural networks, using an\u003cbr\u003earchitecture and a cost function well-adapted to the nature of the provided information. The resulting model is evaluated on an ABX minimal-pair discrimination task and is shown to perform much better (11.8% ABX error\u003cbr\u003erate) than raw speech features (19.6%), not far from a fully supervised baseline (best neural network: 9.2%, HMM-GMM: 11%).","descriptionType":"Abstract"}],"geoLocations":[],"fundingReferences":[],"xml":"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","url":"https://figshare.com/articles/dataset/Phonetics_Embedding_Learning_with_Side_Information/1277821","contentUrl":null,"metadataVersion":1,"schemaVersion":"http://datacite.org/schema/kernel-4","source":"mds","isActive":true,"state":"findable","reason":null,"viewCount":0,"viewsOverTime":[],"downloadCount":0,"downloadsOverTime":[],"referenceCount":0,"citationCount":0,"citationsOverTime":[],"partCount":0,"partOfCount":0,"versionCount":0,"versionOfCount":0,"created":"2014-12-23T18:58:56.000Z","registered":"2014-12-23T18:58:58.000Z","published":"2014","updated":"2020-09-04T19:49:29.000Z"},"relationships":{"client":{"data":{"id":"figshare.ars","type":"clients"}},"provider":{"data":{"id":"otjm","type":"providers"}},"media":{"data":{"id":"10.6084/m9.figshare.1277821","type":"media"}},"references":{"data":[]},"citations":{"data":[]},"parts":{"data":[]},"partOf":{"data":[]},"versions":{"data":[]},"versionOf":{"data":[]}}}}