Matches in SemOpenAlex for { <https://semopenalex.org/work/W4320032114> ?p ?o ?g. }
Showing items 1 to 73 of
73
with 100 items per page.
- W4320032114 abstract "We present a method for visually-grounded spoken term discovery. After training either a HuBERT or wav2vec2.0 model to associate spoken captions with natural images, we show that powerful word segmentation and clustering capability emerges within the model's self-attention heads. Our experiments reveal that this ability is not present to nearly the same extent in the base HuBERT and wav2vec2.0 models, suggesting that the visual grounding task is a crucial component of the word discovery capability we observe. We also evaluate our method on the Buckeye word segmentation and ZeroSpeech spoken term discovery tasks, where we perform on par with or better than currently published methods on several metrics. Code and model weights are available at https://github.com/jasonppy/word-discovery." @default.
- W4320032114 created "2023-02-12" @default.
- W4320032114 creator A5004717608 @default.
- W4320032114 creator A5075735963 @default.
- W4320032114 date "2022-03-28" @default.
- W4320032114 modified "2023-10-18" @default.
- W4320032114 title "Word Discovery in Visually Grounded, Self-Supervised Speech Models" @default.
- W4320032114 doi "https://doi.org/10.48550/arxiv.2203.15081" @default.
- W4320032114 hasPublicationYear "2022" @default.
- W4320032114 type Work @default.
- W4320032114 citedByCount "0" @default.
- W4320032114 crossrefType "posted-content" @default.
- W4320032114 hasAuthorship W4320032114A5004717608 @default.
- W4320032114 hasAuthorship W4320032114A5075735963 @default.
- W4320032114 hasBestOaLocation W43200321141 @default.
- W4320032114 hasConcept C121332964 @default.
- W4320032114 hasConcept C138885662 @default.
- W4320032114 hasConcept C154945302 @default.
- W4320032114 hasConcept C162324750 @default.
- W4320032114 hasConcept C168167062 @default.
- W4320032114 hasConcept C177264268 @default.
- W4320032114 hasConcept C187736073 @default.
- W4320032114 hasConcept C199360897 @default.
- W4320032114 hasConcept C204321447 @default.
- W4320032114 hasConcept C2776760102 @default.
- W4320032114 hasConcept C2780451532 @default.
- W4320032114 hasConcept C28490314 @default.
- W4320032114 hasConcept C41008148 @default.
- W4320032114 hasConcept C41895202 @default.
- W4320032114 hasConcept C61797465 @default.
- W4320032114 hasConcept C62520636 @default.
- W4320032114 hasConcept C73555534 @default.
- W4320032114 hasConcept C89600930 @default.
- W4320032114 hasConcept C90805587 @default.
- W4320032114 hasConcept C97355855 @default.
- W4320032114 hasConcept C98501671 @default.
- W4320032114 hasConceptScore W4320032114C121332964 @default.
- W4320032114 hasConceptScore W4320032114C138885662 @default.
- W4320032114 hasConceptScore W4320032114C154945302 @default.
- W4320032114 hasConceptScore W4320032114C162324750 @default.
- W4320032114 hasConceptScore W4320032114C168167062 @default.
- W4320032114 hasConceptScore W4320032114C177264268 @default.
- W4320032114 hasConceptScore W4320032114C187736073 @default.
- W4320032114 hasConceptScore W4320032114C199360897 @default.
- W4320032114 hasConceptScore W4320032114C204321447 @default.
- W4320032114 hasConceptScore W4320032114C2776760102 @default.
- W4320032114 hasConceptScore W4320032114C2780451532 @default.
- W4320032114 hasConceptScore W4320032114C28490314 @default.
- W4320032114 hasConceptScore W4320032114C41008148 @default.
- W4320032114 hasConceptScore W4320032114C41895202 @default.
- W4320032114 hasConceptScore W4320032114C61797465 @default.
- W4320032114 hasConceptScore W4320032114C62520636 @default.
- W4320032114 hasConceptScore W4320032114C73555534 @default.
- W4320032114 hasConceptScore W4320032114C89600930 @default.
- W4320032114 hasConceptScore W4320032114C90805587 @default.
- W4320032114 hasConceptScore W4320032114C97355855 @default.
- W4320032114 hasConceptScore W4320032114C98501671 @default.
- W4320032114 hasLocation W43200321141 @default.
- W4320032114 hasOpenAccess W4320032114 @default.
- W4320032114 hasPrimaryLocation W43200321141 @default.
- W4320032114 hasRelatedWork W1539050421 @default.
- W4320032114 hasRelatedWork W2072278013 @default.
- W4320032114 hasRelatedWork W2081647779 @default.
- W4320032114 hasRelatedWork W2161919705 @default.
- W4320032114 hasRelatedWork W2357339972 @default.
- W4320032114 hasRelatedWork W2578916128 @default.
- W4320032114 hasRelatedWork W2804033347 @default.
- W4320032114 hasRelatedWork W2883550961 @default.
- W4320032114 hasRelatedWork W2951061418 @default.
- W4320032114 hasRelatedWork W3185852197 @default.
- W4320032114 isParatext "false" @default.
- W4320032114 isRetracted "false" @default.
- W4320032114 workType "article" @default.