Sotto Voce: Federated Speech Recognition with Differential Privacy Guarantees. (arXiv:2207.07816v1 [cs.CR])

Speech data is expensive to collect, and incredibly sensitive to its sources.
It is often the case that organizations independently collect small datasets
for their own use, but often these are not performant for the demands of
machine learning. Organizations could pool these datasets together and jointly
build a strong ASR system; sharing data in the clear, however, comes with
tremendous risk, in terms of intellectual property loss as well as loss of
privacy of the individuals who exist in …

differential privacy privacy recognition speech