FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3106255532> ?p ?o ?g. }

• W3106255532 abstract "Training deep neural networks on well-understood dependencies in speech data can provide new insights into how they learn internal representations. This paper argues that acquisition of speech can be modeled as a dependency between random space and generated speech data in the Generative Adversarial Network architecture and proposes a methodology to uncover the network's internal representations that correspond to phonetic and phonological properties. The Generative Adversarial architecture is uniquely appropriate for modeling phonetic and phonological learning because the network is trained on unannotated raw acoustic data and learning is unsupervised without any language-specific assumptions or pre-assumed levels of abstraction. A Generative Adversarial Network was trained on an allophonic distribution in English, in which voiceless stops surface as aspirated word-initially before stressed vowels, except if preceded by a sibilant [s]. The network successfully learns the allophonic alternation: the network's generated speech signal contains the conditional distribution of aspiration duration. The paper proposes a technique for establishing the network's internal representations that identifies latent variables that correspond to, for example, presence of [s] and its spectral properties. By manipulating these variables, we actively control the presence of [s] and its frication amplitude in the generated outputs. This suggests that the network learns to use latent variables as an approximation of phonetic and phonological representations. Crucially, we observe that the dependencies learned in training extend beyond the training interval, which allows for additional exploration of learning representations. The paper also discusses how the network's architecture and innovative outputs resemble and differ from linguistic behavior in language acquisition, speech disorders, and speech errors, and how well-understood dependencies in speech data can help us interpret how neural networks learn their representations." @default.
• W3106255532 created "2020-11-23" @default.
• W3106255532 creator A5090898601 @default.
• W3106255532 date "2020-07-08" @default.
• W3106255532 modified "2023-10-14" @default.
• W3106255532 title "Generative Adversarial Phonology: Modeling Unsupervised Phonetic and Phonological Learning With Neural Networks" @default.
• W3106255532 cites W1521160348 @default.
• W3106255532 cites W1545920196 @default.
• W3106255532 cites W1576255537 @default.
• W3106255532 cites W1635512741 @default.
• W3106255532 cites W1640915152 @default.
• W3106255532 cites W1861287692 @default.
• W3106255532 cites W1879556487 @default.
• W3106255532 cites W1895119566 @default.
• W3106255532 cites W1937295144 @default.
• W3106255532 cites W1949377791 @default.
• W3106255532 cites W1972092540 @default.
• W3106255532 cites W1976715660 @default.
• W3106255532 cites W1977576460 @default.
• W3106255532 cites W1978224525 @default.
• W3106255532 cites W1980156703 @default.
• W3106255532 cites W1980862600 @default.
• W3106255532 cites W2000823678 @default.
• W3106255532 cites W2005066437 @default.
• W3106255532 cites W2007251877 @default.
• W3106255532 cites W2007564063 @default.
• W3106255532 cites W2011192906 @default.
• W3106255532 cites W2021743383 @default.
• W3106255532 cites W2027785264 @default.
• W3106255532 cites W2028395321 @default.
• W3106255532 cites W2039768054 @default.
• W3106255532 cites W2049287228 @default.
• W3106255532 cites W2050580244 @default.
• W3106255532 cites W2054948443 @default.
• W3106255532 cites W2055722702 @default.
• W3106255532 cites W2065266357 @default.
• W3106255532 cites W2069886770 @default.
• W3106255532 cites W2079207700 @default.
• W3106255532 cites W2079292333 @default.
• W3106255532 cites W2086391558 @default.
• W3106255532 cites W2094786609 @default.
• W3106255532 cites W2099052600 @default.
• W3106255532 cites W2101534803 @default.
• W3106255532 cites W2111476268 @default.
• W3106255532 cites W2118156823 @default.
• W3106255532 cites W2118786898 @default.
• W3106255532 cites W2119160928 @default.
• W3106255532 cites W2120159535 @default.
• W3106255532 cites W2120321299 @default.
• W3106255532 cites W2124938732 @default.
• W3106255532 cites W2129087637 @default.
• W3106255532 cites W2135704565 @default.
• W3106255532 cites W2140676414 @default.
• W3106255532 cites W2142772987 @default.
• W3106255532 cites W2152134037 @default.
• W3106255532 cites W2155721440 @default.
• W3106255532 cites W2159024232 @default.
• W3106255532 cites W2252657604 @default.
• W3106255532 cites W2297002301 @default.
• W3106255532 cites W2404799143 @default.
• W3106255532 cites W2479452957 @default.
• W3106255532 cites W2479658425 @default.
• W3106255532 cites W2483390977 @default.
• W3106255532 cites W2486937529 @default.
• W3106255532 cites W2525364895 @default.
• W3106255532 cites W2602083531 @default.
• W3106255532 cites W2605118633 @default.
• W3106255532 cites W2617453236 @default.
• W3106255532 cites W2794233296 @default.
• W3106255532 cites W2806275854 @default.
• W3106255532 cites W2888482245 @default.
• W3106255532 cites W2898929289 @default.
• W3106255532 cites W2901600249 @default.
• W3106255532 cites W2928787164 @default.
• W3106255532 cites W2963834860 @default.
• W3106255532 cites W2972867623 @default.
• W3106255532 cites W3093872682 @default.
• W3106255532 cites W3105376354 @default.
• W3106255532 cites W3192659128 @default.
• W3106255532 cites W4206592572 @default.
• W3106255532 cites W4210671811 @default.
• W3106255532 cites W4230289889 @default.
• W3106255532 cites W4245961824 @default.
• W3106255532 cites W4246991306 @default.
• W3106255532 cites W4251671532 @default.
• W3106255532 cites W4252588748 @default.
• W3106255532 cites W91681889 @default.
• W3106255532 doi "https://doi.org/10.3389/frai.2020.00044" @default.
• W3106255532 hasPubMedCentralId "https://www.ncbi.nlm.nih.gov/pmc/articles/7861218" @default.
• W3106255532 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/33733161" @default.
• W3106255532 hasPublicationYear "2020" @default.
• W3106255532 type Work @default.
• W3106255532 sameAs 3106255532 @default.
• W3106255532 citedByCount "10" @default.
• W3106255532 countsByYear W31062555322020 @default.
• W3106255532 countsByYear W31062555322021 @default.
• W3106255532 countsByYear W31062555322022 @default.
• W3106255532 countsByYear W31062555322023 @default.
• W3106255532 crossrefType "journal-article" @default.
• W3106255532 hasAuthorship W3106255532A5090898601 @default.

• next