Matches in SemOpenAlex for { <https://semopenalex.org/work/W2980813547> ?p ?o ?g. }
- W2980813547 abstract "Tone is a prosodic feature used to distinguish words in many languages, some of which are endangered and scarcely documented. In this work, we use unsupervised representation learning to identify probable clusters of syllables that share the same phonemic tone. Our method extracts the pitch for each syllable, then trains a convolutional autoencoder to learn a low dimensional representation for each contour. We then apply the mean shift algorithm to cluster tones in high-density regions of the latent space. Furthermore, by feeding the centers of each cluster into the decoder, we produce a prototypical contour that represents each cluster. We apply this method to spoken multi-syllable words in Mandarin Chinese and Cantonese and evaluate how closely our clusters match the ground truth tone categories. Finally, we discuss some difficulties with our approach, including contextual tone variation and allophony effects." @default.
- W2980813547 created "2019-10-25" @default.
- W2980813547 creator A5004641991 @default.
- W2980813547 creator A5048603612 @default.
- W2980813547 creator A5056256317 @default.
- W2980813547 date "2019-10-20" @default.
- W2980813547 modified "2023-09-23" @default.
- W2980813547 title "Representation Learning for Discovering Phonemic Tone Contours" @default.
- W2980813547 cites W1796128977 @default.
- W2980813547 cites W1993799394 @default.
- W2980813547 cites W2016619091 @default.
- W2980813547 cites W2032429609 @default.
- W2980813547 cites W2038784064 @default.
- W2980813547 cites W2067191022 @default.
- W2980813547 cites W2115296722 @default.
- W2980813547 cites W2118680762 @default.
- W2980813547 cites W2137692354 @default.
- W2980813547 cites W2509088535 @default.
- W2980813547 cites W2668277942 @default.
- W2980813547 cites W273093436 @default.
- W2980813547 cites W2786608204 @default.
- W2980813547 cites W2884225676 @default.
- W2980813547 cites W2887260487 @default.
- W2980813547 cites W2899771611 @default.
- W2980813547 cites W2964121744 @default.
- W2980813547 cites W2972442845 @default.
- W2980813547 cites W2994325518 @default.
- W2980813547 cites W3017204694 @default.
- W2980813547 cites W3217769081 @default.
- W2980813547 doi "https://doi.org/10.48550/arxiv.1910.08987" @default.
- W2980813547 hasPublicationYear "2019" @default.
- W2980813547 type Work @default.
- W2980813547 sameAs 2980813547 @default.
- W2980813547 citedByCount "1" @default.
- W2980813547 countsByYear W29808135472019 @default.
- W2980813547 crossrefType "posted-content" @default.
- W2980813547 hasAuthorship W2980813547A5004641991 @default.
- W2980813547 hasAuthorship W2980813547A5048603612 @default.
- W2980813547 hasAuthorship W2980813547A5056256317 @default.
- W2980813547 hasBestOaLocation W29808135471 @default.
- W2980813547 hasConcept C101738243 @default.
- W2980813547 hasConcept C108583219 @default.
- W2980813547 hasConcept C109089402 @default.
- W2980813547 hasConcept C121332964 @default.
- W2980813547 hasConcept C138885662 @default.
- W2980813547 hasConcept C138954614 @default.
- W2980813547 hasConcept C153180895 @default.
- W2980813547 hasConcept C154945302 @default.
- W2980813547 hasConcept C17744445 @default.
- W2980813547 hasConcept C199539241 @default.
- W2980813547 hasConcept C204321447 @default.
- W2980813547 hasConcept C2776359362 @default.
- W2980813547 hasConcept C2776401178 @default.
- W2980813547 hasConcept C2777895490 @default.
- W2980813547 hasConcept C2778334786 @default.
- W2980813547 hasConcept C2780583480 @default.
- W2980813547 hasConcept C28490314 @default.
- W2980813547 hasConcept C41008148 @default.
- W2980813547 hasConcept C41895202 @default.
- W2980813547 hasConcept C44870925 @default.
- W2980813547 hasConcept C59404180 @default.
- W2980813547 hasConcept C94625758 @default.
- W2980813547 hasConceptScore W2980813547C101738243 @default.
- W2980813547 hasConceptScore W2980813547C108583219 @default.
- W2980813547 hasConceptScore W2980813547C109089402 @default.
- W2980813547 hasConceptScore W2980813547C121332964 @default.
- W2980813547 hasConceptScore W2980813547C138885662 @default.
- W2980813547 hasConceptScore W2980813547C138954614 @default.
- W2980813547 hasConceptScore W2980813547C153180895 @default.
- W2980813547 hasConceptScore W2980813547C154945302 @default.
- W2980813547 hasConceptScore W2980813547C17744445 @default.
- W2980813547 hasConceptScore W2980813547C199539241 @default.
- W2980813547 hasConceptScore W2980813547C204321447 @default.
- W2980813547 hasConceptScore W2980813547C2776359362 @default.
- W2980813547 hasConceptScore W2980813547C2776401178 @default.
- W2980813547 hasConceptScore W2980813547C2777895490 @default.
- W2980813547 hasConceptScore W2980813547C2778334786 @default.
- W2980813547 hasConceptScore W2980813547C2780583480 @default.
- W2980813547 hasConceptScore W2980813547C28490314 @default.
- W2980813547 hasConceptScore W2980813547C41008148 @default.
- W2980813547 hasConceptScore W2980813547C41895202 @default.
- W2980813547 hasConceptScore W2980813547C44870925 @default.
- W2980813547 hasConceptScore W2980813547C59404180 @default.
- W2980813547 hasConceptScore W2980813547C94625758 @default.
- W2980813547 hasLocation W29808135471 @default.
- W2980813547 hasOpenAccess W2980813547 @default.
- W2980813547 hasPrimaryLocation W29808135471 @default.
- W2980813547 hasRelatedWork W196866866 @default.
- W2980813547 hasRelatedWork W1982856740 @default.
- W2980813547 hasRelatedWork W2029478385 @default.
- W2980813547 hasRelatedWork W2050631214 @default.
- W2980813547 hasRelatedWork W2104743630 @default.
- W2980813547 hasRelatedWork W2785535669 @default.
- W2980813547 hasRelatedWork W2980813547 @default.
- W2980813547 hasRelatedWork W3037952526 @default.
- W2980813547 hasRelatedWork W44671426 @default.
- W2980813547 hasRelatedWork W2520379491 @default.
- W2980813547 isParatext "false" @default.
- W2980813547 isRetracted "false" @default.
- W2980813547 magId "2980813547" @default.