spoken-to-signed-translation
spoken-to-signed-translation copied to clipboard
Incorporate Gloss Sense Disambiguation
Our current text-to-gloss translation can lead to ambiguity in the generated glosses. We propose enhancing our system by incorporating gloss sense disambiguation to better capture the intended meaning of the input text.
To illustrate how big of a problem this is, here are four translations for `KLEIN'
https://signsuisse.sgb-fss.ch/fileadmin/signsuisse_ressources/videos/EB019CAA-D768-D07E-571E219A5F6AA5AB.mp4 https://signsuisse.sgb-fss.ch/fileadmin/signsuisse_ressources/videos/FC9A32C5-C1B1-9B67-BB8C75F7039B568D.mp4 https://signsuisse.sgb-fss.ch/fileadmin/signsuisse_ressources/videos/79B5C9E9-FFBD-7DD2-D579FB9D6E5BC384.mp4 https://signsuisse.sgb-fss.ch/fileadmin/signsuisse_ressources/videos/2171343E-AEF1-76DD-E213221C22CB0261.mp4
And it is unclear how to choose.