FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313136502> ?p ?o ?g. }

• W4313136502 abstract "Over the last decade, deep-learning methods have been gradually incorporated into conventional automatic speech recognition (ASR) frameworks to create acoustic, pronunciation, and language models. Although it led to significant improvements in ASRs' recognition accuracy, due to their hard constraints related to hardware requirements (e.g., computing power and memory usage), it is unclear if such approaches are the most computationally- and energy-efficient options for embedded ASR applications. Reservoir computing (RC) models (e.g., echo state networks (ESNs) and liquid state machines (LSMs)), on the other hand, have been proven inexpensive to train, have vastly fewer parameters, and are compatible with emergent hardware technologies. However, their performance in speech processing tasks is relatively inferior to that of the deep-learning-based models. To enhance the accuracy of the RC in ASR applications, we propose heterogeneous single and multi-layer ESNs to create non-linear transformations of the inputs that capture temporal context at different scales. To test our models, we performed a speech recognition task on the Farsdat Persian dataset. Since, to the best of our knowledge, standard RC has not yet been employed to conduct any Persian ASR tasks, we also trained conventional single-layer and deep ESNs to provide baselines for comparison. Besides, we compared the RC performance with a standard long-short-term memory (LSTM) model. Heterogeneous RC models (1) show improved performance to the standard RC models; (2) perform on par in terms of recognition accuracy with the LSTM, and (3) reduce the training time considerably." @default.
• W4313136502 created "2023-01-06" @default.
• W4313136502 creator A5001991567 @default.
• W4313136502 creator A5082705031 @default.
• W4313136502 creator A5089128852 @default.
• W4313136502 date "2022-07-18" @default.
• W4313136502 modified "2023-09-28" @default.
• W4313136502 title "Heterogeneous Reservoir Computing Models for Persian Speech Recognition" @default.
• W4313136502 cites W1609149460 @default.
• W4313136502 cites W1919069 @default.
• W4313136502 cites W1983239292 @default.
• W4313136502 cites W2001263627 @default.
• W4313136502 cites W2011975175 @default.
• W4313136502 cites W2022011789 @default.
• W4313136502 cites W2037024114 @default.
• W4313136502 cites W2064675550 @default.
• W4313136502 cites W2079735306 @default.
• W4313136502 cites W2095168618 @default.
• W4313136502 cites W2103179919 @default.
• W4313136502 cites W2118706537 @default.
• W4313136502 cites W2143612262 @default.
• W4313136502 cites W2147768505 @default.
• W4313136502 cites W2157331557 @default.
• W4313136502 cites W2159549127 @default.
• W4313136502 cites W2160815625 @default.
• W4313136502 cites W2467101772 @default.
• W4313136502 cites W2469050391 @default.
• W4313136502 cites W2608997467 @default.
• W4313136502 cites W2771720152 @default.
• W4313136502 cites W2943513641 @default.
• W4313136502 cites W2946257168 @default.
• W4313136502 cites W2969267557 @default.
• W4313136502 cites W2975924256 @default.
• W4313136502 cites W3097538676 @default.
• W4313136502 cites W3119055764 @default.
• W4313136502 cites W3120556073 @default.
• W4313136502 cites W3136954484 @default.
• W4313136502 cites W3155334984 @default.
• W4313136502 cites W3162345085 @default.
• W4313136502 doi "https://doi.org/10.1109/ijcnn55064.2022.9892570" @default.
• W4313136502 hasPublicationYear "2022" @default.
• W4313136502 type Work @default.
• W4313136502 citedByCount "1" @default.
• W4313136502 countsByYear W43131365022023 @default.
• W4313136502 crossrefType "proceedings-article" @default.
• W4313136502 hasAuthorship W4313136502A5001991567 @default.
• W4313136502 hasAuthorship W4313136502A5082705031 @default.
• W4313136502 hasAuthorship W4313136502A5089128852 @default.
• W4313136502 hasBestOaLocation W43131365022 @default.
• W4313136502 hasConcept C108583219 @default.
• W4313136502 hasConcept C133488467 @default.
• W4313136502 hasConcept C135796866 @default.
• W4313136502 hasConcept C138885662 @default.
• W4313136502 hasConcept C147168706 @default.
• W4313136502 hasConcept C151730666 @default.
• W4313136502 hasConcept C154945302 @default.
• W4313136502 hasConcept C162324750 @default.
• W4313136502 hasConcept C187736073 @default.
• W4313136502 hasConcept C2779343474 @default.
• W4313136502 hasConcept C2780451532 @default.
• W4313136502 hasConcept C2780844864 @default.
• W4313136502 hasConcept C28490314 @default.
• W4313136502 hasConcept C41008148 @default.
• W4313136502 hasConcept C41895202 @default.
• W4313136502 hasConcept C50644808 @default.
• W4313136502 hasConcept C86803240 @default.
• W4313136502 hasConceptScore W4313136502C108583219 @default.
• W4313136502 hasConceptScore W4313136502C133488467 @default.
• W4313136502 hasConceptScore W4313136502C135796866 @default.
• W4313136502 hasConceptScore W4313136502C138885662 @default.
• W4313136502 hasConceptScore W4313136502C147168706 @default.
• W4313136502 hasConceptScore W4313136502C151730666 @default.
• W4313136502 hasConceptScore W4313136502C154945302 @default.
• W4313136502 hasConceptScore W4313136502C162324750 @default.
• W4313136502 hasConceptScore W4313136502C187736073 @default.
• W4313136502 hasConceptScore W4313136502C2779343474 @default.
• W4313136502 hasConceptScore W4313136502C2780451532 @default.
• W4313136502 hasConceptScore W4313136502C2780844864 @default.
• W4313136502 hasConceptScore W4313136502C28490314 @default.
• W4313136502 hasConceptScore W4313136502C41008148 @default.
• W4313136502 hasConceptScore W4313136502C41895202 @default.
• W4313136502 hasConceptScore W4313136502C50644808 @default.
• W4313136502 hasConceptScore W4313136502C86803240 @default.
• W4313136502 hasFunder F4320320879 @default.
• W4313136502 hasLocation W43131365021 @default.
• W4313136502 hasLocation W43131365022 @default.
• W4313136502 hasOpenAccess W4313136502 @default.
• W4313136502 hasPrimaryLocation W43131365021 @default.
• W4313136502 hasRelatedWork W2126887587 @default.
• W4313136502 hasRelatedWork W2731899572 @default.
• W4313136502 hasRelatedWork W2939353110 @default.
• W4313136502 hasRelatedWork W3009238340 @default.
• W4313136502 hasRelatedWork W3215138031 @default.
• W4313136502 hasRelatedWork W4297924634 @default.
• W4313136502 hasRelatedWork W4315783664 @default.
• W4313136502 hasRelatedWork W4321369474 @default.
• W4313136502 hasRelatedWork W4327774331 @default.
• W4313136502 hasRelatedWork W4360585206 @default.
• W4313136502 isParatext "false" @default.
• W4313136502 isRetracted "false" @default.

• next