| dc.contributor.author | Kolesau, Aliaksei | |
| dc.contributor.author | Šešok, Dmitrij | |
| dc.date.accessioned | 2023-09-18T20:45:00Z | |
| dc.date.available | 2023-09-18T20:45:00Z | |
| dc.date.issued | 2021 | |
| dc.identifier.isbn | 9781665449281 | |
| dc.identifier.uri | https://etalpykla.vilniustech.lt/handle/123456789/152313 | |
| dc.description.abstract | In this paper we propose a practical method to improve the false rejection rate of a voice-activated system. The users of such systems tend to repeat the activation word in the case of the false rejection. In our method we modify a decoder of a phoneme-based voice-activated system for detecting such repeats which allows to catch 14.6% of previously undetected cases, while not increasing the false alarm rate, and 79% while adding less than 1% of false alarm rate. We use the fact that the repeat of the activation word consists of twice as many phonemes than the single pronunciation which allows more reliable detection. | eng |
| dc.format | PDF | |
| dc.format.extent | p. 1-2 | |
| dc.format.medium | tekstas / txt | |
| dc.language.iso | eng | |
| dc.relation.isreferencedby | IEEE Xplore | |
| dc.relation.isreferencedby | Scopus | |
| dc.rights | Prieinamas tik institucijos(-ų) intranete | |
| dc.source.uri | https://ieeexplore.ieee.org/document/9431481 | |
| dc.source.uri | https://talpykla.elaba.lt/elaba-fedora/objects/elaba:98769021/datastreams/MAIN/content | |
| dc.title | Detecting wake-word repeats in voice-activated systems | |
| dc.type | Straipsnis konferencijos darbų leidinyje Scopus DB / Paper in conference publication in Scopus DB | |
| dcterms.references | 11 | |
| dc.type.pubtype | P1b - Straipsnis konferencijos darbų leidinyje Scopus DB / Article in conference proceedings Scopus DB | |
| dc.contributor.institution | Vilniaus Gedimino technikos universitetas | |
| dc.contributor.faculty | Fundamentinių mokslų fakultetas / Faculty of Fundamental Sciences | |
| dc.subject.researchfield | T 007 - Informatikos inžinerija / Informatics engineering | |
| dc.subject.vgtuprioritizedfields | IK0303 - Dirbtinio intelekto ir sprendimų priėmimo sistemos / Artificial intelligence and decision support systems | |
| dc.subject.ltspecializations | L106 - Transportas, logistika ir informacinės ir ryšių technologijos (IRT) / Transport, logistic and information and communication technologies | |
| dc.subject.en | voice-activated system | |
| dc.subject.en | keyword spotting | |
| dc.subject.en | time delay neural network | |
| dcterms.sourcetitle | 2021 IEEE Open Conference of Electrical, Electronic and Information Sciences (eStream), 22 April 2021, Vilnius, Lithuania / organized by: Vilnius Gediminas Technical University | |
| dc.identifier.eissn | 2690-8506 | |
| dc.publisher.name | IEEE | |
| dc.publisher.city | Piscataway, NJ | |
| dc.identifier.doi | 10.1109/eStream53087.2021.9431481 | |
| dc.identifier.elaba | 98769021 | |
