The University of Southampton
University of Southampton Institutional Repository

Practical audio system design for private speech reproduction

Practical audio system design for private speech reproduction
Practical audio system design for private speech reproduction
Multi-zone sound field control allows individuals to listen to personalised audio content whilst sharing a physical space. Applications of this technology include home entertainment, audio reproduction in public spaces such as museums, shops or exhibitions, and providing areas where the privacy of sensitive communication can be safeguarded without the need for physical barriers. The problem of transmitting a speech signal to a single listener and reducing the intelligibility of that signal elsewhere is the focus of the present thesis. The motivation behind the presented experiments and simulations is to identify the practical trade-offs that must be considered in the design of these Speech Privacy Control systems. Conventional personal audio systems use loudspeaker array processing to produce a bright zone for the intended user of the system and a dark zone where silence is desired. However, established performance metrics and system optimisation techniques do not necessarily yield privacy for the target listener, as attenuated speech may remain intelligible within the dark zone. A system is proposed that focusses a synthetic masking signal into the dark zone to selectively reduce the intelligibility of the leaked speech. Privacy is ensured by adjusting the masker to meet predefined constraints on the speech intelligibility in each zone. This design methodology utilises information from speech intelligibility tests and subjective preference evaluations in order to improve the utility and acceptability of such systems for all nearby listeners. In addition to the design of the masking signal, the performance of a speech privacy control system is affected by the loudspeaker array design and the location of the listening zones. These effects are explored using experimental measurements of a loudspeaker array in a room, and the results are used to select two system configurations for additional evaluation using listening tests. The perceived performance of a system is also affected by the surrounding acoustic environment, notably due to reverberation and background noise, which may change over time. The effects of room reverberation are investigated using image source simulations and acoustical measurements within a room, and the performance is evaluated in terms of the achievable level of acoustic contrast, the difference in speech intelligibility between zones, and the masking signal levels that are required to achieve privacy. A proposal is made to further enhance privacy by combining the effects of background noise and artificial masking signals. This method reduces the level of acoustic contrast that is required to achieve a given level of privacy, compared to the case where the masking is provided by the background noise alone.
University of Southampton
Wallace, Daniel
ef3e070e-d641-48ac-8e5b-fe083131ee86
Wallace, Daniel
ef3e070e-d641-48ac-8e5b-fe083131ee86
Cheer, Jordan
8e452f50-4c7d-4d4e-913a-34015e99b9dc

Wallace, Daniel (2020) Practical audio system design for private speech reproduction. University of Southampton, Doctoral Thesis, 216pp.

Record type: Thesis (Doctoral)

Abstract

Multi-zone sound field control allows individuals to listen to personalised audio content whilst sharing a physical space. Applications of this technology include home entertainment, audio reproduction in public spaces such as museums, shops or exhibitions, and providing areas where the privacy of sensitive communication can be safeguarded without the need for physical barriers. The problem of transmitting a speech signal to a single listener and reducing the intelligibility of that signal elsewhere is the focus of the present thesis. The motivation behind the presented experiments and simulations is to identify the practical trade-offs that must be considered in the design of these Speech Privacy Control systems. Conventional personal audio systems use loudspeaker array processing to produce a bright zone for the intended user of the system and a dark zone where silence is desired. However, established performance metrics and system optimisation techniques do not necessarily yield privacy for the target listener, as attenuated speech may remain intelligible within the dark zone. A system is proposed that focusses a synthetic masking signal into the dark zone to selectively reduce the intelligibility of the leaked speech. Privacy is ensured by adjusting the masker to meet predefined constraints on the speech intelligibility in each zone. This design methodology utilises information from speech intelligibility tests and subjective preference evaluations in order to improve the utility and acceptability of such systems for all nearby listeners. In addition to the design of the masking signal, the performance of a speech privacy control system is affected by the loudspeaker array design and the location of the listening zones. These effects are explored using experimental measurements of a loudspeaker array in a room, and the results are used to select two system configurations for additional evaluation using listening tests. The perceived performance of a system is also affected by the surrounding acoustic environment, notably due to reverberation and background noise, which may change over time. The effects of room reverberation are investigated using image source simulations and acoustical measurements within a room, and the performance is evaluated in terms of the achievable level of acoustic contrast, the difference in speech intelligibility between zones, and the masking signal levels that are required to achieve privacy. A proposal is made to further enhance privacy by combining the effects of background noise and artificial masking signals. This method reduces the level of acoustic contrast that is required to achieve a given level of privacy, compared to the case where the masking is provided by the background noise alone.

Text
Practical Audio System Design for Private Speech Reproduction - Version of Record
Available under License University of Southampton Thesis Licence.
Download (21MB)
Text
Thesis DRAFT 20200728 - Other
Restricted to Repository staff only
Text
Practical Audio System Design for Private Speech Reproduction
Available under License University of Southampton Thesis Licence.
Download (22MB)
Text
Permission to deposit thesis - form
Restricted to Repository staff only

More information

Published date: September 2020

Identifiers

Local EPrints ID: 448173
URI: http://eprints.soton.ac.uk/id/eprint/448173
PURE UUID: 9d2f8001-cb47-4934-85e6-932e73c46c59
ORCID for Daniel Wallace: ORCID iD orcid.org/0000-0003-0212-5395
ORCID for Jordan Cheer: ORCID iD orcid.org/0000-0002-0552-5506

Catalogue record

Date deposited: 14 Apr 2021 16:31
Last modified: 18 Mar 2024 03:17

Export record

Contributors

Author: Daniel Wallace ORCID iD
Thesis advisor: Jordan Cheer ORCID iD

Download statistics

Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.

View more statistics

Atom RSS 1.0 RSS 2.0

Contact ePrints Soton: eprints@soton.ac.uk

ePrints Soton supports OAI 2.0 with a base URL of http://eprints.soton.ac.uk/cgi/oai2

This repository has been built using EPrints software, developed at the University of Southampton, but available to everyone to use.

We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.

×