Matches in SemOpenAlex for { <https://semopenalex.org/work/W3136804610> ?p ?o ?g. }
Showing items 1 to 77 of
77
with 100 items per page.
- W3136804610 abstract "Achieving good accuracy in Deep Learning (DL) models for Wake Word Engines (WWE) strongly depends on using the right type of dataset in training. Important WWE dataset dimensions are the number of wake words, the ratio of wake words to non-wake words, the composition of non-wake words and the augmentation techniques applied to the training dataset. In this paper, we make two crucial and first-of-a-kind contributions in improving the WWE performance: (a) Extensively studying and proving the effectiveness of three potent audio augmentation techniques that are previously not studied or less-studied for WWE and (b) Providing clear insight into the effect of wake-word to non-wake word ratio and the composition of the latter, specifically the impact of using non-wake words uttered by the speakers who also utter the wake words (vs) using non-wake words that are mined from elsewhere,We prove the effectiveness of our recipe of: (a) Data augmentation with Ambient Sound addition, Reverberation, Echo and Speed along with SpecAugment and (b) A non-wake word composition of Common Voice plus Freesound audio slices at a wake word to non-wake word ratio of 1:15 and 1:20 respectively. We show that the WWE model trained with our recipe for ‘Computer' exhibits a False Reject Rate (FRR) at 1 False Alarm (FA) per hour of 0%, 5.6%, 0.4%, and 0.7% under the four test environments trialed, whereas, the standard technique trained model exhibits an FRR of 1.1%, 18.3%, 3.7%, and 3.7%. Similar results are observed for wake word ‘Crystal'." @default.
- W3136804610 created "2021-03-29" @default.
- W3136804610 creator A5005614987 @default.
- W3136804610 creator A5040352022 @default.
- W3136804610 creator A5060799266 @default.
- W3136804610 creator A5069403056 @default.
- W3136804610 creator A5079851149 @default.
- W3136804610 date "2020-12-10" @default.
- W3136804610 modified "2023-10-18" @default.
- W3136804610 title "Recipe for Creating a Highly Accurate Wake Word Engine" @default.
- W3136804610 cites W1496120315 @default.
- W3136804610 cites W2034940213 @default.
- W3136804610 cites W2406222150 @default.
- W3136804610 cites W2407023693 @default.
- W3136804610 cites W2407080277 @default.
- W3136804610 cites W2591789009 @default.
- W3136804610 cites W2617258110 @default.
- W3136804610 cites W2696967604 @default.
- W3136804610 cites W2774918761 @default.
- W3136804610 cites W2887528618 @default.
- W3136804610 cites W2888797456 @default.
- W3136804610 cites W2891138528 @default.
- W3136804610 cites W2936774411 @default.
- W3136804610 cites W2963414149 @default.
- W3136804610 cites W2972541648 @default.
- W3136804610 cites W2972572074 @default.
- W3136804610 cites W2975761646 @default.
- W3136804610 cites W3007380371 @default.
- W3136804610 cites W3008587939 @default.
- W3136804610 cites W3015995734 @default.
- W3136804610 cites W3098773154 @default.
- W3136804610 doi "https://doi.org/10.1109/bigdata50022.2020.9378193" @default.
- W3136804610 hasPublicationYear "2020" @default.
- W3136804610 type Work @default.
- W3136804610 sameAs 3136804610 @default.
- W3136804610 citedByCount "0" @default.
- W3136804610 crossrefType "proceedings-article" @default.
- W3136804610 hasAuthorship W3136804610A5005614987 @default.
- W3136804610 hasAuthorship W3136804610A5040352022 @default.
- W3136804610 hasAuthorship W3136804610A5060799266 @default.
- W3136804610 hasAuthorship W3136804610A5069403056 @default.
- W3136804610 hasAuthorship W3136804610A5079851149 @default.
- W3136804610 hasConcept C127413603 @default.
- W3136804610 hasConcept C146978453 @default.
- W3136804610 hasConcept C154945302 @default.
- W3136804610 hasConcept C2524010 @default.
- W3136804610 hasConcept C28490314 @default.
- W3136804610 hasConcept C33923547 @default.
- W3136804610 hasConcept C41008148 @default.
- W3136804610 hasConcept C48939323 @default.
- W3136804610 hasConcept C90805587 @default.
- W3136804610 hasConceptScore W3136804610C127413603 @default.
- W3136804610 hasConceptScore W3136804610C146978453 @default.
- W3136804610 hasConceptScore W3136804610C154945302 @default.
- W3136804610 hasConceptScore W3136804610C2524010 @default.
- W3136804610 hasConceptScore W3136804610C28490314 @default.
- W3136804610 hasConceptScore W3136804610C33923547 @default.
- W3136804610 hasConceptScore W3136804610C41008148 @default.
- W3136804610 hasConceptScore W3136804610C48939323 @default.
- W3136804610 hasConceptScore W3136804610C90805587 @default.
- W3136804610 hasLocation W31368046101 @default.
- W3136804610 hasOpenAccess W3136804610 @default.
- W3136804610 hasPrimaryLocation W31368046101 @default.
- W3136804610 hasRelatedWork W10910819 @default.
- W3136804610 hasRelatedWork W13270548 @default.
- W3136804610 hasRelatedWork W13718537 @default.
- W3136804610 hasRelatedWork W14230040 @default.
- W3136804610 hasRelatedWork W2864082 @default.
- W3136804610 hasRelatedWork W5949257 @default.
- W3136804610 hasRelatedWork W7296353 @default.
- W3136804610 hasRelatedWork W816280 @default.
- W3136804610 hasRelatedWork W8738421 @default.
- W3136804610 hasRelatedWork W9638664 @default.
- W3136804610 isParatext "false" @default.
- W3136804610 isRetracted "false" @default.
- W3136804610 magId "3136804610" @default.
- W3136804610 workType "article" @default.