Using privacy-transformed speech in the automatic speech recognition acoustic model training
Résumé
Automatic Speech Recognition (ASR) requires huge amounts of real user speech data to reach state-of-the-art performance. However, speech data conveys sensitive speaker attributes like identity that can be inferred and exploited for malicious purposes. Therefore, there is a interest in collection of the anonymized speech data that is processed by some voice conversion method. In this paper we evaluate one of voice conversion methods on Latvian speech data and also investigate if privacy-transformed data can be used to improve ASR acoustic models. Results show effectiveness of voice conversion against state-of-the-art speaker verification models on Latvian speech and effectiveness of using privacy-transformed data in ASR training.
Origine | Fichiers produits par l'(les) auteur(s) |
---|
Loading...