Matches in SemOpenAlex for { <https://semopenalex.org/work/W2975842671> ?p ?o ?g. }
Showing items 1 to 74 of
74
with 100 items per page.
- W2975842671 abstract "In order to build language technologies for majority of the languages, it is important to leverage the resources available in public domain on the internet - commonly referred to as `Found Data'. However, such data is characterized by the presence of non-standard, non-trivial variations. For instance, speech resources found on the internet have non-speech content, such as music. Therefore, speech recognition and speech synthesis models need to be robust to such variations. In this work, we present an analysis to show that it is important to disentangle the latent causal factors of variation in the original data to accomplish these tasks. Based on this, we present approaches to disentangle such variations from the data using Latent Stochastic Models. Specifically, we present a method to split the latent prior space into continuous representations of dominant speech modes present in the magnitude spectra of audio signals. We propose a completely unsupervised approach using multinode latent space variational autoencoders (VAE). We show that the constraints on the latent space of a VAE can be in-fact used to separate speech and music, independent of the language of the speech. This paper also analytically presents the requirement on the number of latent variables for the task based on distribution of the speech data." @default.
- W2975842671 created "2019-10-03" @default.
- W2975842671 creator A5036498144 @default.
- W2975842671 creator A5046364646 @default.
- W2975842671 creator A5073104107 @default.
- W2975842671 date "2019-09-25" @default.
- W2975842671 modified "2023-09-26" @default.
- W2975842671 title "Disentangling Speech and Non-Speech Components for Building Robust Acoustic Models from Found Data." @default.
- W2975842671 cites W1533145941 @default.
- W2975842671 cites W1731081199 @default.
- W2975842671 cites W2022668263 @default.
- W2975842671 cites W2094562799 @default.
- W2975842671 cites W2164098335 @default.
- W2975842671 cites W2509006290 @default.
- W2975842671 cites W2514828952 @default.
- W2975842671 cites W2802304149 @default.
- W2975842671 cites W2899379230 @default.
- W2975842671 cites W2907262790 @default.
- W2975842671 cites W2937197076 @default.
- W2975842671 cites W2963204961 @default.
- W2975842671 hasPublicationYear "2019" @default.
- W2975842671 type Work @default.
- W2975842671 sameAs 2975842671 @default.
- W2975842671 citedByCount "2" @default.
- W2975842671 countsByYear W29758426712020 @default.
- W2975842671 crossrefType "posted-content" @default.
- W2975842671 hasAuthorship W2975842671A5036498144 @default.
- W2975842671 hasAuthorship W2975842671A5046364646 @default.
- W2975842671 hasAuthorship W2975842671A5073104107 @default.
- W2975842671 hasConcept C111919701 @default.
- W2975842671 hasConcept C153083717 @default.
- W2975842671 hasConcept C154945302 @default.
- W2975842671 hasConcept C170133592 @default.
- W2975842671 hasConcept C2778572836 @default.
- W2975842671 hasConcept C28490314 @default.
- W2975842671 hasConcept C41008148 @default.
- W2975842671 hasConcept C51167844 @default.
- W2975842671 hasConcept C61328038 @default.
- W2975842671 hasConceptScore W2975842671C111919701 @default.
- W2975842671 hasConceptScore W2975842671C153083717 @default.
- W2975842671 hasConceptScore W2975842671C154945302 @default.
- W2975842671 hasConceptScore W2975842671C170133592 @default.
- W2975842671 hasConceptScore W2975842671C2778572836 @default.
- W2975842671 hasConceptScore W2975842671C28490314 @default.
- W2975842671 hasConceptScore W2975842671C41008148 @default.
- W2975842671 hasConceptScore W2975842671C51167844 @default.
- W2975842671 hasConceptScore W2975842671C61328038 @default.
- W2975842671 hasLocation W29758426711 @default.
- W2975842671 hasOpenAccess W2975842671 @default.
- W2975842671 hasPrimaryLocation W29758426711 @default.
- W2975842671 hasRelatedWork W1841395947 @default.
- W2975842671 hasRelatedWork W1852631781 @default.
- W2975842671 hasRelatedWork W2023952145 @default.
- W2975842671 hasRelatedWork W2139450919 @default.
- W2975842671 hasRelatedWork W2157940937 @default.
- W2975842671 hasRelatedWork W2225932399 @default.
- W2975842671 hasRelatedWork W2256811946 @default.
- W2975842671 hasRelatedWork W2514426326 @default.
- W2975842671 hasRelatedWork W2563686518 @default.
- W2975842671 hasRelatedWork W2737108017 @default.
- W2975842671 hasRelatedWork W2907262790 @default.
- W2975842671 hasRelatedWork W2935938411 @default.
- W2975842671 hasRelatedWork W2964058423 @default.
- W2975842671 hasRelatedWork W3020570669 @default.
- W2975842671 hasRelatedWork W3127721277 @default.
- W2975842671 hasRelatedWork W3146320432 @default.
- W2975842671 hasRelatedWork W3155909695 @default.
- W2975842671 hasRelatedWork W3175161143 @default.
- W2975842671 hasRelatedWork W46110029 @default.
- W2975842671 hasRelatedWork W289577468 @default.
- W2975842671 isParatext "false" @default.
- W2975842671 isRetracted "false" @default.
- W2975842671 magId "2975842671" @default.
- W2975842671 workType "article" @default.