New benchmark measures gender bias in speech translation systems
http://feedproxy.google.com/~r/venturebeat/SZYF/~3/g_VDlUvKvNo/
A preprint paper published by University of Trento researchers proposes a benchmark — MuST-SHE — to evaluate whether speech translation systems fed textual data are constrained by the fact that sentences sometimes omit gender identity clues. It’s the assertion of the coauthors that these systems can and do exhibit gender bias, » ….