3 papers
eess.AS2019
End-to-end architectures for ASR-free spoken language understanding
Elisavet Palogiannidi, Ioannis Gkinis, George Mastrapas +2
Spoken Language Understanding (SLU) is the problem of extracting the meaning from speech utterances. It is typically addressed as a two-step problem, where an Automatic Speech Reco…
eess.AS2019
Privacy-Preserving Speaker Recognition with Cohort Score Normalisation
Andreas Nautsch, Jose Patino, Amos Treiber +5
In many voice biometrics applications there is a requirement to preserve privacy, not least because of the recently enforced General Data Protection Regulation (GDPR). Though progr…
cs.CV2019
Self-supervised speaker embeddings
Themos Stafylakis, Johan Rohdin, Oldrich Plchot +2
Contrary to i-vectors, speaker embeddings such as x-vectors are incapable of leveraging unlabelled utterances, due to the classification loss over training speakers. In this paper,…