Matches in SemOpenAlex for { <https://semopenalex.org/work/W3139534224> ?p ?o ?g. }
- W3139534224 abstract "Acoustic word embeddings (AWEs) are fixed-dimensional representations of variable-length speech segments. For zero-resource languages where labelled data is not available, one AWE approach is to use unsupervised autoencoder-based recurrent models. Another recent approach is to use multilingual transfer: a supervised AWE model is trained on several well-resourced languages and then applied to an unseen zero-resource language. We consider how a recent contrastive learning loss can be used in both the purely unsupervised and multilingual transfer settings. Firstly, we show that terms from an unsupervised term discovery system can be used for contrastive self-supervision, resulting in improvements over previous unsupervised monolingual AWE models. Secondly, we consider how multilingual AWE models can be adapted to a specific zero-resource language using discovered terms. We find that self-supervised contrastive adaptation outperforms adapted multilingual correspondence autoencoder and Siamese AWE models, giving the best overall results in a word discrimination task on six zero-resource languages." @default.
- W3139534224 created "2021-03-29" @default.
- W3139534224 creator A5040305929 @default.
- W3139534224 creator A5053159197 @default.
- W3139534224 creator A5078855278 @default.
- W3139534224 date "2021-03-19" @default.
- W3139534224 modified "2023-09-27" @default.
- W3139534224 title "Acoustic word embeddings for zero-resource languages using self-supervised contrastive learning and multilingual adaptation." @default.
- W3139534224 cites W1532499126 @default.
- W3139534224 cites W1577418252 @default.
- W3139534224 cites W2025482506 @default.
- W3139534224 cites W2057007397 @default.
- W3139534224 cites W2059652594 @default.
- W3139534224 cites W2084534958 @default.
- W3139534224 cites W2106053110 @default.
- W3139534224 cites W2114347655 @default.
- W3139534224 cites W2165698076 @default.
- W3139534224 cites W2190506272 @default.
- W3139534224 cites W2291770225 @default.
- W3139534224 cites W2296681920 @default.
- W3139534224 cites W2321533354 @default.
- W3139534224 cites W2407151108 @default.
- W3139534224 cites W2550241133 @default.
- W3139534224 cites W2566587499 @default.
- W3139534224 cites W2578392894 @default.
- W3139534224 cites W2598634450 @default.
- W3139534224 cites W2802557066 @default.
- W3139534224 cites W2884305338 @default.
- W3139534224 cites W2889313720 @default.
- W3139534224 cites W2899518769 @default.
- W3139534224 cites W2927191280 @default.
- W3139534224 cites W2932675979 @default.
- W3139534224 cites W2935542736 @default.
- W3139534224 cites W2962736743 @default.
- W3139534224 cites W2962742544 @default.
- W3139534224 cites W2963259843 @default.
- W3139534224 cites W2963425185 @default.
- W3139534224 cites W2963879199 @default.
- W3139534224 cites W2964102148 @default.
- W3139534224 cites W2964121744 @default.
- W3139534224 cites W2964169922 @default.
- W3139534224 cites W2973026522 @default.
- W3139534224 cites W2979736514 @default.
- W3139534224 cites W2980988419 @default.
- W3139534224 cites W2991213871 @default.
- W3139534224 cites W2995489995 @default.
- W3139534224 cites W2996383576 @default.
- W3139534224 cites W3003875258 @default.
- W3139534224 cites W3005680577 @default.
- W3139534224 cites W3008444411 @default.
- W3139534224 cites W3015213852 @default.
- W3139534224 cites W3015325583 @default.
- W3139534224 cites W3015356564 @default.
- W3139534224 cites W3026574827 @default.
- W3139534224 cites W3033831118 @default.
- W3139534224 cites W3037580942 @default.
- W3139534224 cites W3046603443 @default.
- W3139534224 cites W3047836878 @default.
- W3139534224 cites W3096196861 @default.
- W3139534224 cites W3096656254 @default.
- W3139534224 cites W3105242324 @default.
- W3139534224 cites W3162468843 @default.
- W3139534224 cites W343636949 @default.
- W3139534224 hasPublicationYear "2021" @default.
- W3139534224 type Work @default.
- W3139534224 sameAs 3139534224 @default.
- W3139534224 citedByCount "0" @default.
- W3139534224 crossrefType "posted-content" @default.
- W3139534224 hasAuthorship W3139534224A5040305929 @default.
- W3139534224 hasAuthorship W3139534224A5053159197 @default.
- W3139534224 hasAuthorship W3139534224A5078855278 @default.
- W3139534224 hasConcept C101738243 @default.
- W3139534224 hasConcept C108583219 @default.
- W3139534224 hasConcept C138885662 @default.
- W3139534224 hasConcept C139807058 @default.
- W3139534224 hasConcept C150899416 @default.
- W3139534224 hasConcept C154945302 @default.
- W3139534224 hasConcept C15744967 @default.
- W3139534224 hasConcept C162324750 @default.
- W3139534224 hasConcept C169760540 @default.
- W3139534224 hasConcept C187736073 @default.
- W3139534224 hasConcept C204321447 @default.
- W3139534224 hasConcept C206345919 @default.
- W3139534224 hasConcept C2780451532 @default.
- W3139534224 hasConcept C2780813799 @default.
- W3139534224 hasConcept C31258907 @default.
- W3139534224 hasConcept C41008148 @default.
- W3139534224 hasConcept C41895202 @default.
- W3139534224 hasConcept C8038995 @default.
- W3139534224 hasConcept C90805587 @default.
- W3139534224 hasConceptScore W3139534224C101738243 @default.
- W3139534224 hasConceptScore W3139534224C108583219 @default.
- W3139534224 hasConceptScore W3139534224C138885662 @default.
- W3139534224 hasConceptScore W3139534224C139807058 @default.
- W3139534224 hasConceptScore W3139534224C150899416 @default.
- W3139534224 hasConceptScore W3139534224C154945302 @default.
- W3139534224 hasConceptScore W3139534224C15744967 @default.
- W3139534224 hasConceptScore W3139534224C162324750 @default.
- W3139534224 hasConceptScore W3139534224C169760540 @default.
- W3139534224 hasConceptScore W3139534224C187736073 @default.