Matches in SemOpenAlex for { <https://semopenalex.org/work/W2512885752> ?p ?o ?g. }
Showing items 1 to 72 of
72
with 100 items per page.
- W2512885752 abstract "This paper focuses on the robust recognition of nonlinearly distorted speech. We have reported (Seps et al.,2014) that hybrid acoustic models based on a combination of Hidden Markov Models and Deep Neural Networks(HMM-DNNs) are better suited to this task than conventional HMMs utilizing Gaussian Mixture Models(HMM-GMMs). To further improve recognition accuracy, this paper investigates the possibility of combiningthe modeling power of deep neural networks with the adaptation to given acoustic conditions. For thispurpose, the deep neural networks are utilized to produce bottleneck coefficients / features (BNC). The BNCsare subsequently used for training of HMM-GMM based acoustic models and then adapted using ConstrainedMaximum Likelihood Linear Regression (CMLLR). Our results obtained for three types of nonlinear distortionsand three types of input features show that the adapted BNC-based system (a) outperforms HMM-DNNacoustic models in the case of strong compression and (b) yields comparable performance for speech affectedby nonlinear amplification in the analog domain." @default.
- W2512885752 created "2016-09-16" @default.
- W2512885752 creator A5004206733 @default.
- W2512885752 creator A5007129369 @default.
- W2512885752 creator A5033391650 @default.
- W2512885752 creator A5075473112 @default.
- W2512885752 date "2016-01-01" @default.
- W2512885752 modified "2023-09-25" @default.
- W2512885752 title "Study on the Use and Adaptation of Bottleneck Features for Robust Speech Recognition of Nonlinearly Distorted Speech" @default.
- W2512885752 doi "https://doi.org/10.5220/0005955500650071" @default.
- W2512885752 hasPublicationYear "2016" @default.
- W2512885752 type Work @default.
- W2512885752 sameAs 2512885752 @default.
- W2512885752 citedByCount "1" @default.
- W2512885752 countsByYear W25128857522021 @default.
- W2512885752 crossrefType "proceedings-article" @default.
- W2512885752 hasAuthorship W2512885752A5004206733 @default.
- W2512885752 hasAuthorship W2512885752A5007129369 @default.
- W2512885752 hasAuthorship W2512885752A5033391650 @default.
- W2512885752 hasAuthorship W2512885752A5075473112 @default.
- W2512885752 hasBestOaLocation W25128857521 @default.
- W2512885752 hasConcept C149635348 @default.
- W2512885752 hasConcept C153180895 @default.
- W2512885752 hasConcept C154945302 @default.
- W2512885752 hasConcept C155635449 @default.
- W2512885752 hasConcept C23224414 @default.
- W2512885752 hasConcept C2780513914 @default.
- W2512885752 hasConcept C28490314 @default.
- W2512885752 hasConcept C2984842247 @default.
- W2512885752 hasConcept C41008148 @default.
- W2512885752 hasConcept C50644808 @default.
- W2512885752 hasConcept C61224824 @default.
- W2512885752 hasConcept C61328038 @default.
- W2512885752 hasConceptScore W2512885752C149635348 @default.
- W2512885752 hasConceptScore W2512885752C153180895 @default.
- W2512885752 hasConceptScore W2512885752C154945302 @default.
- W2512885752 hasConceptScore W2512885752C155635449 @default.
- W2512885752 hasConceptScore W2512885752C23224414 @default.
- W2512885752 hasConceptScore W2512885752C2780513914 @default.
- W2512885752 hasConceptScore W2512885752C28490314 @default.
- W2512885752 hasConceptScore W2512885752C2984842247 @default.
- W2512885752 hasConceptScore W2512885752C41008148 @default.
- W2512885752 hasConceptScore W2512885752C50644808 @default.
- W2512885752 hasConceptScore W2512885752C61224824 @default.
- W2512885752 hasConceptScore W2512885752C61328038 @default.
- W2512885752 hasLocation W25128857521 @default.
- W2512885752 hasOpenAccess W2512885752 @default.
- W2512885752 hasPrimaryLocation W25128857521 @default.
- W2512885752 hasRelatedWork W2005689542 @default.
- W2512885752 hasRelatedWork W2029947361 @default.
- W2512885752 hasRelatedWork W2237224976 @default.
- W2512885752 hasRelatedWork W2314175981 @default.
- W2512885752 hasRelatedWork W2397728357 @default.
- W2512885752 hasRelatedWork W2405505124 @default.
- W2512885752 hasRelatedWork W2406310406 @default.
- W2512885752 hasRelatedWork W2561072122 @default.
- W2512885752 hasRelatedWork W2747374697 @default.
- W2512885752 hasRelatedWork W2753802668 @default.
- W2512885752 hasRelatedWork W2755245168 @default.
- W2512885752 hasRelatedWork W2891980359 @default.
- W2512885752 hasRelatedWork W2897067191 @default.
- W2512885752 hasRelatedWork W2901279995 @default.
- W2512885752 hasRelatedWork W2913014027 @default.
- W2512885752 hasRelatedWork W2945616452 @default.
- W2512885752 hasRelatedWork W2963470561 @default.
- W2512885752 hasRelatedWork W2963597470 @default.
- W2512885752 hasRelatedWork W2966220259 @default.
- W2512885752 hasRelatedWork W3202469742 @default.
- W2512885752 isParatext "false" @default.
- W2512885752 isRetracted "false" @default.
- W2512885752 magId "2512885752" @default.
- W2512885752 workType "article" @default.