| fullrecord |
[{"key": "dc.contributor.advisor", "value": "K\u00e4rkk\u00e4inen, Tommi", "language": "", "element": "contributor", "qualifier": "advisor", "schema": "dc"}, {"key": "dc.contributor.advisor", "value": "H\u00e4m\u00e4l\u00e4inen, Joonas", "language": "", "element": "contributor", "qualifier": "advisor", "schema": "dc"}, {"key": "dc.contributor.author", "value": "Jokinen, Ville", "language": "", "element": "contributor", "qualifier": "author", "schema": "dc"}, {"key": "dc.date.accessioned", "value": "2021-06-28T13:30:50Z", "language": null, "element": "date", "qualifier": "accessioned", "schema": "dc"}, {"key": "dc.date.available", "value": "2021-06-28T13:30:50Z", "language": null, "element": "date", "qualifier": "available", "schema": "dc"}, {"key": "dc.date.issued", "value": "2021", "language": "", "element": "date", "qualifier": "issued", "schema": "dc"}, {"key": "dc.identifier.uri", "value": "https://jyx.jyu.fi/handle/123456789/76866", "language": null, "element": "identifier", "qualifier": "uri", "schema": "dc"}, {"key": "dc.description.abstract", "value": "Tutkielman tavoitteena on vertailla uusimpia koneoppimiseen pohjautuvia menetelmi\u00e4 puhujan tunnistamiseen v\u00e4h\u00e4isell\u00e4 datan m\u00e4\u00e4r\u00e4ll\u00e4. Puhujan tunnistamisessa tavoitteena on tunnistaa eri puhujat \u00e4\u00e4nidatasta, sen k\u00e4ytt\u00f6tarkoituksiin sis\u00e4ltyy mm. puhujan diarioiminen ja biometrinen tunnistus \u00e4\u00e4nen avulla. Tutkielma rajoittuu puhujan tapaukseen, jossa k\u00e4ytett\u00e4viss\u00e4 on kaksi lyhytt\u00e4 nauhoitetta, joko yhdelt\u00e4 tai kahdelta, ennest\u00e4\u00e4n tuntemattomalta puhujalta. Joiden pohjalta pyrit\u00e4\u00e4n tunnistamaan, sis\u00e4lt\u00e4v\u00e4tk\u00f6 nauhoitteet puhetta samalta puhujalta. Lis\u00e4ksi tutkielmassa tutkitaan Englanninkielisell\u00e4 puheella koulutettujen neuroverkkojen tarkkuutta Suomenkieliseen puheeseen sovellettuna. Johon kehitet\u00e4\u00e4n sopiva datasetti Suomenkielisen puhekorpuksen pohjalta.\n\nTutkielman tulokset osoittavat uusimpien menetelmien suoriutuvan erinomaisesti. Vaikkakin parhaiden tuloksien saavuttaminen osoittautui vaativan enemm\u00e4n koulutusdataa kuin mit\u00e4 tutkielmassa k\u00e4ytet\u00e4\u00e4n. Menetelm\u00e4t yleistyv\u00e4t hyvin my\u00f6s suomenkieliselle puheelle siit\u00e4 huolimatta, ett\u00e4 koulutuksessa k\u00e4ytettiin vain englanninkielist\u00e4 puhetta. Lis\u00e4ksi tuloksien pohjalta tehd\u00e4\u00e4n mielenkiintoisia huomioita vertailuun valittujen muuttujien osalta, joita k\u00e4ytet\u00e4\u00e4n neuroverkkojen koulutuksessa. Vertailussa oli menetelmien lis\u00e4ksi koulutusdatan puhujien m\u00e4\u00e4r\u00e4, puhe esimerkkin pituus ja \u00e4\u00e4nidatan augmentointi.", "language": "fi", "element": "description", "qualifier": "abstract", "schema": "dc"}, {"key": "dc.description.abstract", "value": "This thesis sets out to compare recent methods in speaker recognition, from a small amount of data. Speaker recognition aims to distinguish speakers from within audio data containing speech, the use cases include for example speaker diarization and voice biometric authentication. The scope is limited to identification, two samples from one or two distinct previously unknown speakers are provided. With the aim being to identify whether the two samples are spoken by the same speaker. Additionally, the accuracy of networks trained on English speech on Finnish speech is also measured. For which a new dataset, suitable for benchmarking speaker recognition, consisting of Finnish speech was developed from an existing speech recognition dataset.\n\nThe results show that the latest methods perform very well. However, to achieve the best results it is apparent that more training data is required, than what was used in this thesis. The methods generalized to Finnish speech, despite being trained with English speech. Additionally, interesting observations are made regarding the parameters chosen for training. In addition to comparing different methods, the effects of different number of speakers used for training, various sample lengths and data augmentation are also compared.", "language": "en", "element": "description", "qualifier": "abstract", "schema": "dc"}, {"key": "dc.description.provenance", "value": "Submitted by Miia Hakanen (mihakane@jyu.fi) on 2021-06-28T13:30:50Z\nNo. of bitstreams: 0", "language": "en", "element": "description", "qualifier": "provenance", "schema": "dc"}, {"key": "dc.description.provenance", "value": "Made available in DSpace on 2021-06-28T13:30:50Z (GMT). No. of bitstreams: 0\n Previous issue date: 2021", "language": "en", "element": "description", "qualifier": "provenance", "schema": "dc"}, {"key": "dc.format.extent", "value": "69", "language": "", "element": "format", "qualifier": "extent", "schema": "dc"}, {"key": "dc.format.mimetype", "value": "application/pdf", "language": null, "element": "format", "qualifier": "mimetype", "schema": "dc"}, {"key": "dc.language.iso", "value": "eng", "language": null, "element": "language", "qualifier": "iso", "schema": "dc"}, {"key": "dc.rights", "value": "In Copyright", "language": "en", "element": "rights", "qualifier": null, "schema": "dc"}, {"key": "dc.subject.other", "value": "speaker identification", "language": "", "element": "subject", "qualifier": "other", "schema": "dc"}, {"key": "dc.subject.other", "value": "few-shot learning", "language": "", "element": "subject", "qualifier": "other", "schema": "dc"}, {"key": "dc.title", "value": "Few-shot learning for speaker recognition", "language": "", "element": "title", "qualifier": null, "schema": "dc"}, {"key": "dc.type", "value": "master thesis", "language": null, "element": "type", "qualifier": null, "schema": "dc"}, {"key": "dc.identifier.urn", "value": "URN:NBN:fi:jyu-202106284056", "language": "", "element": "identifier", "qualifier": "urn", "schema": "dc"}, {"key": "dc.type.ontasot", "value": "Pro gradu -tutkielma", "language": "fi", "element": "type", "qualifier": "ontasot", "schema": "dc"}, {"key": "dc.type.ontasot", "value": "Master\u2019s thesis", "language": "en", "element": "type", "qualifier": "ontasot", "schema": "dc"}, {"key": "dc.contributor.faculty", "value": "Informaatioteknologian tiedekunta", "language": "fi", "element": "contributor", "qualifier": "faculty", "schema": "dc"}, {"key": "dc.contributor.faculty", "value": "Faculty of Information Technology", "language": "en", "element": "contributor", "qualifier": "faculty", "schema": "dc"}, {"key": "dc.contributor.department", "value": "Informaatioteknologia", "language": "fi", "element": "contributor", "qualifier": "department", "schema": "dc"}, {"key": "dc.contributor.department", "value": "Information Technology", "language": "en", "element": "contributor", "qualifier": "department", "schema": "dc"}, {"key": "dc.contributor.organization", "value": "Jyv\u00e4skyl\u00e4n yliopisto", "language": "fi", "element": "contributor", "qualifier": "organization", "schema": "dc"}, {"key": "dc.contributor.organization", "value": "University of Jyv\u00e4skyl\u00e4", "language": "en", "element": "contributor", "qualifier": "organization", "schema": "dc"}, {"key": "dc.subject.discipline", "value": "Tietotekniikka", "language": "fi", "element": "subject", "qualifier": "discipline", "schema": "dc"}, {"key": "dc.subject.discipline", "value": "Mathematical Information Technology", "language": "en", "element": "subject", "qualifier": "discipline", "schema": "dc"}, {"key": "yvv.contractresearch.funding", "value": "0", "language": "", "element": "contractresearch", "qualifier": "funding", "schema": "yvv"}, {"key": "dc.type.coar", "value": "http://purl.org/coar/resource_type/c_bdcc", "language": null, "element": "type", "qualifier": "coar", "schema": "dc"}, {"key": "dc.rights.accesslevel", "value": "openAccess", "language": null, "element": "rights", "qualifier": "accesslevel", "schema": "dc"}, {"key": "dc.type.publication", "value": "masterThesis", "language": null, "element": "type", "qualifier": "publication", "schema": "dc"}, {"key": "dc.subject.oppiainekoodi", "value": "602", "language": "", "element": "subject", "qualifier": "oppiainekoodi", "schema": "dc"}, {"key": "dc.subject.yso", "value": "koneoppiminen", "language": null, "element": "subject", "qualifier": "yso", "schema": "dc"}, {"key": "dc.subject.yso", "value": "puhujantunnistus", "language": null, "element": "subject", "qualifier": "yso", "schema": "dc"}, {"key": "dc.subject.yso", "value": "neuroverkot", "language": null, "element": "subject", "qualifier": "yso", "schema": "dc"}, {"key": "dc.subject.yso", "value": "machine learning", "language": null, "element": "subject", "qualifier": "yso", "schema": "dc"}, {"key": "dc.subject.yso", "value": "speaker recognition", "language": null, "element": "subject", "qualifier": "yso", "schema": "dc"}, {"key": "dc.subject.yso", "value": "neural networks (information technology)", "language": null, "element": "subject", "qualifier": "yso", "schema": "dc"}, {"key": "dc.format.content", "value": "fulltext", "language": null, "element": "format", "qualifier": "content", "schema": "dc"}, {"key": "dc.rights.url", "value": "https://rightsstatements.org/page/InC/1.0/", "language": null, "element": "rights", "qualifier": "url", "schema": "dc"}, {"key": "dc.type.okm", "value": "G2", "language": null, "element": "type", "qualifier": "okm", "schema": "dc"}]
|