Privacy-preserving Query-by-Example Speech Search

Abstract

This paper investigates a new privacy-preserving paradigm for the task of Query-by-Example Speech Search using Secure Binary Embeddings, a hashing method that converts vector data to bit strings through a combination of random projections followed by banded quantization. The proposed method allows performing spoken query search in an encrypted domain, by analyzing ciphered information computed from the original recordings. Unlike other hashing techniques, the embeddings allow the computation of the distance between vectors that are close enough, but are not perfect matches. This paper shows how these hashes can be combined with Dynamic Time Warping based on posterior derived features to perform secure speech search. Experiments performed on a sub-set of the SpeechDat Portuguese corpus showed that the proposed privacy-preserving system obtains similar results to its non-private counterpart.

Patient Privacy in Paralinguistic Tasks

“…This scheme allows information to be leaked if the corresponding SBE hashes of two vectors are close enough, while preserving information-theoretic security. It has been applied to several speech processing tasks such as speaker verification [3] and query-by-example speech search [4]. An extension of SBE, Secure Modular Hashing (SMH) [5], allows the user to control the extension of the information leakage, thus enabling a trade-off between accuracy and security.…”