Matches in SemOpenAlex for { <https://semopenalex.org/work/W4377371484> ?p ?o ?g. }
Showing items 1 to 75 of
75
with 100 items per page.
- W4377371484 abstract "In this paper, we show that representations capturing syllabic units emerge when training a self-supervised speech model with a visually-grounded training objective. We demonstrate that a nearly identical model architecture (HuBERT) trained with a masked language modeling loss does not exhibit this same ability, suggesting that the visual grounding objective is responsible for the emergence of this phenomenon. We propose the use of a minimum cut algorithm to automatically predict syllable boundaries in speech, followed by a 2-stage clustering method to group identical syllables together. We show that our model not only outperforms a state-of-the-art syllabic segmentation method on the language it was trained on (English), but also generalizes in a zero-shot fashion to Estonian. Finally, we show that the same model is capable of zero-shot generalization for a word segmentation task on 4 other languages from the Zerospeech Challenge, in some cases beating the previous state-of-the-art." @default.
- W4377371484 created "2023-05-23" @default.
- W4377371484 creator A5004717608 @default.
- W4377371484 creator A5016518233 @default.
- W4377371484 creator A5029566548 @default.
- W4377371484 creator A5075735963 @default.
- W4377371484 creator A5085086690 @default.
- W4377371484 date "2023-05-19" @default.
- W4377371484 modified "2023-09-29" @default.
- W4377371484 title "Syllable Discovery and Cross-Lingual Generalization in a Visually Grounded, Self-Supervised Speech Model" @default.
- W4377371484 doi "https://doi.org/10.48550/arxiv.2305.11435" @default.
- W4377371484 hasPublicationYear "2023" @default.
- W4377371484 type Work @default.
- W4377371484 citedByCount "0" @default.
- W4377371484 crossrefType "posted-content" @default.
- W4377371484 hasAuthorship W4377371484A5004717608 @default.
- W4377371484 hasAuthorship W4377371484A5016518233 @default.
- W4377371484 hasAuthorship W4377371484A5029566548 @default.
- W4377371484 hasAuthorship W4377371484A5075735963 @default.
- W4377371484 hasAuthorship W4377371484A5085086690 @default.
- W4377371484 hasBestOaLocation W43773714841 @default.
- W4377371484 hasConcept C109089402 @default.
- W4377371484 hasConcept C134306372 @default.
- W4377371484 hasConcept C137293760 @default.
- W4377371484 hasConcept C138885662 @default.
- W4377371484 hasConcept C154945302 @default.
- W4377371484 hasConcept C162324750 @default.
- W4377371484 hasConcept C177148314 @default.
- W4377371484 hasConcept C187736073 @default.
- W4377371484 hasConcept C194051139 @default.
- W4377371484 hasConcept C204321447 @default.
- W4377371484 hasConcept C2780451532 @default.
- W4377371484 hasConcept C2780813799 @default.
- W4377371484 hasConcept C28490314 @default.
- W4377371484 hasConcept C33923547 @default.
- W4377371484 hasConcept C41008148 @default.
- W4377371484 hasConcept C41895202 @default.
- W4377371484 hasConcept C73555534 @default.
- W4377371484 hasConcept C89600930 @default.
- W4377371484 hasConcept C90805587 @default.
- W4377371484 hasConceptScore W4377371484C109089402 @default.
- W4377371484 hasConceptScore W4377371484C134306372 @default.
- W4377371484 hasConceptScore W4377371484C137293760 @default.
- W4377371484 hasConceptScore W4377371484C138885662 @default.
- W4377371484 hasConceptScore W4377371484C154945302 @default.
- W4377371484 hasConceptScore W4377371484C162324750 @default.
- W4377371484 hasConceptScore W4377371484C177148314 @default.
- W4377371484 hasConceptScore W4377371484C187736073 @default.
- W4377371484 hasConceptScore W4377371484C194051139 @default.
- W4377371484 hasConceptScore W4377371484C204321447 @default.
- W4377371484 hasConceptScore W4377371484C2780451532 @default.
- W4377371484 hasConceptScore W4377371484C2780813799 @default.
- W4377371484 hasConceptScore W4377371484C28490314 @default.
- W4377371484 hasConceptScore W4377371484C33923547 @default.
- W4377371484 hasConceptScore W4377371484C41008148 @default.
- W4377371484 hasConceptScore W4377371484C41895202 @default.
- W4377371484 hasConceptScore W4377371484C73555534 @default.
- W4377371484 hasConceptScore W4377371484C89600930 @default.
- W4377371484 hasConceptScore W4377371484C90805587 @default.
- W4377371484 hasLocation W43773714841 @default.
- W4377371484 hasOpenAccess W4377371484 @default.
- W4377371484 hasPrimaryLocation W43773714841 @default.
- W4377371484 hasRelatedWork W1491515786 @default.
- W4377371484 hasRelatedWork W1995315771 @default.
- W4377371484 hasRelatedWork W2035773033 @default.
- W4377371484 hasRelatedWork W2159815177 @default.
- W4377371484 hasRelatedWork W2399529871 @default.
- W4377371484 hasRelatedWork W2480617478 @default.
- W4377371484 hasRelatedWork W4238890879 @default.
- W4377371484 hasRelatedWork W4245442210 @default.
- W4377371484 hasRelatedWork W2185854508 @default.
- W4377371484 hasRelatedWork W3120992332 @default.
- W4377371484 isParatext "false" @default.
- W4377371484 isRetracted "false" @default.
- W4377371484 workType "article" @default.