Privacy-Preserving Machine Learning for Speech Processing

by Pathak, Manas A.

Type Book

Rate 0.0 /5

Details

Files

Reviews

Download

Title Statements

Full title :Privacy-Preserving Machine Learning for Speech Processing / Manas A. Pathak.

Summary

This thesis discusses the privacy issues in speech-based applications, including biometric authentication, surveillance, and external speech processing services. Manas A. Pathak presents solutions for privacy-preserving speech processing applications such as speaker verification, speaker identification, and speech recognition. The thesis introduces tools from cryptography and machine learning and current techniques for improving the efficiency and scalability of the presented solutions, as well as experiments with prototype implementations of the solutions for execution time and accuracy on standardized speech datasets. Using the framework proposed may make it possible for a surveillance agency to listen for a known terrorist, without being able to hear conversation from non-targeted, innocent civilians.

Edition Statements

Publication Date:2013

Publication Statements

Publisher(s):Springer
Place of Publication:New York
Language:eng
Publication Date:2013

Notes

"Doctoral Thesis accepted by Carnegie Mellon University, USA"--Title page.

The Physical Object

Description:xvii, 141 pages : : illustrations (some color) ; : 25 cm.

Edition Identifiers

ISBN:9781461446385 (alk. paper), 1461446384 (alk. paper)
DOI:10.1007/978-1-4614-4639-2
oclc:on1353300973
Dewey Decimal:621.382

Machine generated contents note: pt. I Introduction -- 1.Thesis Overview -- 1.1.Motivation -- 1.2.Thesis Statement -- 1.3.Summary of Contributions -- 1.4.Thesis Organization -- References -- 2.Speech Processing Background -- 2.1.Tools and Techniques -- 2.1.1.Signal Parameterization -- 2.1.2.Gaussian Mixture Models -- 2.1.3.Hidden Markov Models -- 2.2.Speaker Identification and Verification -- 2.2.1.Modeling Speech -- 2.2.2.Model Adaptation -- 2.2.3.Supervectors with LSH -- 2.2.4.Reconstructing Data from LSH Keys -- 2.3.Speech Recognition -- References -- 3.Privacy Background -- 3.1.What is Privacy? -- 3.1.1.Definitions -- 3.1.2.Related Concepts -- 3.1.3.Privacy-Preserving Applications -- 3.1.4.Privacy-Preserving Computation in this Thesis -- 3.2.Secure Multiparty Computation -- 3.2.1.Protocol Assumptions -- 3.2.2.Adversarial Behavior -- 3.2.3.Privacy Definitions: Ideal Model and Real Model -- 3.2.4.Encryption -- 3.2.5.Masking --

Note continued: 3.2.6.Zero-Knowledge Proofs and Threshold Cryptosystems -- 3.2.7.Oblivious Transfer -- 3.2.8.Related Work on SMC Protocols for Machine Learning -- 3.3.Differential Privacy -- 3.3.1.Exponential Mechanism -- 3.3.2.Related Work on Differentia…

Subjects

Communications Engineering, Networks Data structures (Computer science) Data Structures, Cryptology and Informa… engineering Machine learning Power Electronics, Electrical Machines … Production of electric energy or power Signal, Image and Speech Processing Speech processing systems TECHNOLOGY & ENGINEERING -- Telecommuni… Telecommunication