Full metadata record
DC pole | Hodnota | Jazyk |
---|---|---|
dc.contributor.author | Švec, Jan | |
dc.contributor.author | Šmídl, Luboš | |
dc.contributor.author | Psutka, Josef | |
dc.contributor.author | Pražák, Aleš | |
dc.date.accessioned | 2022-03-28T10:00:28Z | - |
dc.date.available | 2022-03-28T10:00:28Z | - |
dc.date.issued | 2021 | |
dc.identifier.citation | ŠVEC, J. ŠMÍDL, L. PSUTKA, J. PRAŽÁK, A. Spoken Term Detection and Relevance Score Estimation Using Dot-Product of Pronunciation Embeddings. In Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH. Red Hook, NY: International Speech Communication Association, 2021. s. 851-855. ISBN: 978-1-71383-690-2 , ISSN: 2308-457X | cs |
dc.identifier.isbn | 978-1-71383-690-2 | |
dc.identifier.issn | 2308-457X | |
dc.identifier.uri | 2-s2.0-85119207187 | |
dc.identifier.uri | http://hdl.handle.net/11025/47251 | |
dc.format | 5 s. | cs |
dc.format.mimetype | application/pdf | |
dc.language.iso | en | en |
dc.publisher | International Speech Communication Association | en |
dc.relation.ispartofseries | Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH | en |
dc.rights | Plný text není přístupný. | cs |
dc.rights | © ISCA | en |
dc.title | Spoken Term Detection and Relevance Score Estimation Using Dot-Product of Pronunciation Embeddings | en |
dc.type | konferenční příspěvek | cs |
dc.type | ConferenceObject | en |
dc.rights.access | closedAccess | en |
dc.type.version | publishedVersion | en |
dc.description.abstract-translated | The paper describes a novel approach to Spoken Term Detection (STD) in large spoken archives using deep LSTM networks. The work is based on the previous approach of using Siamese neural networks for STD and naturally extends it to directly localize a spoken term and estimate its relevance score. The phoneme confusion network generated by a phoneme recognizer is processed by the deep LSTM network which projects each segment of the confusion network into an embedding space. The searched term is projected into the same embedding space using another deep LSTM network. The relevance score is then computed using a simple dot-product in the embedding space and calibrated using a sigmoid function to predict the probability of occurrence. The location of the searched term is then estimated from the sequence of output probabilities. The deep LSTM networks are trained in a self-supervised manner from paired recognition hypotheses on word and phoneme levels. The method is experimentally evaluated on MALACH data in English and Czech languages. | en |
dc.subject.translated | spoken term detection | en |
dc.subject.translated | relevance-score estimation | en |
dc.subject.translated | speech embeddings | en |
dc.identifier.doi | 10.21437/Interspeech.2021-1704 | |
dc.type.status | Peer-reviewed | en |
dc.identifier.obd | 43933416 | |
dc.project.ID | VJ01010108/Robustní zpracování nahrávek pro operativu a bezpečnost | cs |
dc.project.ID | 90140/Velká výzkumná infrastruktura_(J) - e-INFRA CZ | cs |
Vyskytuje se v kolekcích: | Konferenční příspěvky / Conference Papers (KKY) OBD |
Soubory připojené k záznamu:
Soubor | Velikost | Formát | |
---|---|---|---|
svec21_interspeech.pdf | 307,53 kB | Adobe PDF | Zobrazit/otevřít Vyžádat kopii |
Použijte tento identifikátor k citaci nebo jako odkaz na tento záznam:
http://hdl.handle.net/11025/47251
Všechny záznamy v DSpace jsou chráněny autorskými právy, všechna práva vyhrazena.