Need addition to the research paper as per Prof comments “Please make sure that you add data to your work and add all the details to your methodology as y
Need addition to the research paper as per Prof comments “Please make sure that you add data to your work and add all the details to your methodology as you go forward. As of now, methodology is in the first and basic stage of the research.”Attaching the research paper for your reference. Running head: ANONYMIZING VOICE INPUT IN MOBILE DEVICES
Anonymizing Voice Input in Mobile Devices
Munish Verma
Harrisburg University of Science and Technology
In Fulfillment of ISEM 699
June 10, 2019
1
ANONYMIZING VOICE INPUT IN MOBILE DEVICES
2
Abstract
This paper aims to reveal the possible solution to the security concern of the interaction between
humans and computers, through an intermediate system between users and the cloud, which
creates user anonymity by sanitizing the voice data of users before they get sent to the cloud for
speech recognition. The technique adopts a two-phase sanitization model, i.e., an attack resistant
voice conversion mechanism and evolution-based keyword substitution. The two phases are all
performed in a user’s mobile device while still ensuring the cloud-supported speech recognition
service is accurate and usable. I estimate that this can minimize the probability of a user’s voice
identified by up to 85%.
Keywords: Voice recognition, cybersecurity, human-computer interaction, HCI,
anonymizing.
ANONYMIZING VOICE INPUT IN MOBILE DEVICES
3
Table of Contents
Abstract …………………………………………………………………………………………………………………………. 2
Table of Contents ……………………………………………………………………………………………………………. 3
Introduction ……………………………………………………………………………………………………………………. 5
Background ………………………………………………………………………………………………………………… 5
Basics of human-computer interaction ……………………………………………………………………………. 6
The motivation behind the creation of voice anonymity……………………………………………………. 8
Growth of voice interactions and sanitization ………………………………………………………………….. 9
Potential benefits of voice sanitization………………………………………………………………………….. 10
Problem Statement …………………………………………………………………………………………………….. 10
Purpose Statement ……………………………………………………………………………………………………… 12
Research Questions and Objectives ……………………………………………………………………………… 13
Theoretical Framework ………………………………………………………………………………………………. 13
Hypothesis ………………………………………………………………………………………………………………… 16
Significance of Study …………………………………………………………………………………………………. 16
Definition of Terms ……………………………………………………………………………………………………. 17
Ethical Considerations………………………………………………………………………………………………… 17
Delimitation………………………………………………………………………………………………………………. 18
Limitations ……………………………………………………………………………………………………………….. 18
Literature Review………………………………………………………………………………………………………….. 18
ANONYMIZING VOICE INPUT IN MOBILE DEVICES
4
Privacy and Security Risks ………………………………………………………………………………………….. 19
Anonymizing voice input ……………………………………………………………………………………………. 20
Anonymizing voice input in mobile devices system architecture ………………………………………… 21
Proposed Research Design and Methodology …………………………………………………………………… 23
Research Worldview ………………………………………………………………………………………………….. 23
Research Design ………………………………………………………………………………………………………… 23
Research Strategy ………………………………………………………………………………………………………. 23
Method of Data Collection ………………………………………………………………………………………….. 24
Purpose of Data Analysis ……………………………………………………………………………………………. 24
Population…………………………………………………………………………………………………………………. 25
Sample ……………………………………………………………………………………………………………………… 25
Proposed Schedule ………………………………………………………………………………………………………… 25
References ……………………………………………………………………………………………………………………. 26
ANONYMIZING VOICE INPUT IN MOBILE DEVICES
5
Introduction
Background
Human-computer interaction (HCI) refers to the study of peoples interaction with
computers and the extent to which machines are or aren’t developed for interacting with humans
successfully (Fernandez et al., 2016). Historically, excluding a few, most computer system
developers failed to pay enough attention to ease of use of computers. Arguably, even today,
some computer users would say that computer makers pay not enough attention to ensuring their
products are user-friendly. Computer system developers may, however, defend themselves by
illustrating the extreme complexity of designing and making computer products and that the
demand for services rendered by computers has always beaten the need for ease of use (Glodek
et al., 2015). Mobile devices nowadays use an input class that offers users access to features such
as a touchscreen, geographical input and accelerometer, and even voice input. Through voice
input, user experience has greatly improved by allowing them to operate their devices handsfree. A technology key to voice input is speech recognition, which often gets outsourced to the
cloud for optimal performance. Outsourcing to the cloud, however, comes with its risks since the
privacy of a user can easily be compromised by matching a voice to a specific user, using speech
recognition to learn their sensitive input content and further profiling users based on such content
(Hwang & Chen, 2017).
When you take a look at hands-free communication, the applications of voice input in
mobile devices are broad especially in areas such as voice search, e.g. Microsoft Bing and
Google, keyboard applications, e.g. Microsoft Sougou, Google, Swift key and in artificial
intelligence personal assistants, e.g. Google assistant, Apple’s Siri and Amazon Echo. Voice
input tremendously improves user experience and lives by freeing users’ hands and attention
ANONYMIZING VOICE INPUT IN MOBILE DEVICES
from time-consuming typing on the usually small screen of mobile devices and is also the
primary means through which visually impaired people achieve human-computer interaction.
Speech-to-text conversion, also known as speech recognition, is a crucial technology behind
successful voice input. Through it, users’ vocal information in the form of spoken language gets
recognized then translated into text. However, due to limited resources on mobile devices,
speech recognition usually gets outsourced to cloud servers for optimal efficiency and accuracy
(Glodek et al., 2015).
Most voice input service providers collect users’ speech records since once the voice
datasets get correlated to the speakers’ identities, plenty of private information that describes the
individual can get revealed. Such Personally Identifiable Information should never get disclosed
on the cloud (Hwang & Chen, 2017). At the moment, it is achievable by leveraging on
anonymous networks like Tor. However, due to inefficient voice sanitization, the voice records
can still get de-anonymized using linkage attacks, which are possible due to the varying voice
biometrics of different individuals.
Basics of human-computer interaction
Human-computer interaction as mentioned in the previous section revolves around the
study that seeks to gather information about the potential behavior that the population exhibits
when it comes to dealing or working with computers among other technologically advanced
devices. Some of the most common methods that the population deals or interacts with include
mobile phones, smartphones, PDAs, and laptops, among others. All these devices play a critical
role in defining the ultimate platform that allows people to undertake numerous tasks. Some of
the functions that the population may benefit from using these devices include communication.
Communication forms a critical part in the overall interaction between people and computers,
6
ANONYMIZING VOICE INPUT IN MOBILE DEVICES
7
among other methods. In addition to that, global interactions allow the populations to address
some of the most common issues that may arise in the course of operations such as security and
speed.
Despite the fact that the advancement of the technology seeks to provide a platform that
makes it easy for the population to undertake communications faster and at higher speeds, it is
worth noting that the use of the resulting gadgets seeks to provide a secure platform for passing
information from one point to the other.
The motivation behind the creation of the several security mechanisms to come from the
understanding that the past and traditional strategies contained diverse challenges which range
from inadequate security protocols to low speeds and hence reduced dependence. However, with
the adoption of the latest technologies, people today can pass information from one point to the
other securely. The improved security of the data in this context basis its arguments from the
notion that the advancement of the levels of technology leads to the discovery of flaws which
perpetrators and third parties can exploit.
The continued discovery of these flaws makes it possible to develop strategies meant to
provide the ultimate solution to the underlying security challenges that the population may
encounter when interacting with the mentioned devices. Some of the issues that arise when it
comes to the implementation of the respective technologies include undermining of the concepts
of security. The advancement of technology makes it easy for third parties to improve on their
ways of intercepting communications between or amongst multiple parties. Sometimes, the
safety of the data is not as paramount as the privacy of the communicating parties.
ANONYMIZING VOICE INPUT IN MOBILE DEVICES
8
There exist multiple ways and technologies that can help in the provision of a secure
communication platform between two parties however, the introduction of the concepts of
voice masks. The primary aim of the use of voice masks in the communications between parties
revolves around the desire to provide a secure mechanism for the protection of the privacy of the
respective stakeholders. In this context, two communicating parties using mobile devices or
smartphone may benefit from the services offered by voice masking due to the fact that it
provides a platform that prevents the third parties from identifying or pinpointing the true or real
identity of the affected members.
The motivation behind the creation of voice anonymity
As mentioned above, the advancement in the levels of technology gave rise to the
occurrence of various improvements in the manner people interact with their mobile devices and
computers. Experts argue that speech recognition stands out as one of the latest technologies that
seem to revolutionize the manner in which people interact with their mobile devices. In addition
to that, the voice-based interactions between people and machines, in the end, play a critical part
in the improvement of the overall user experience. User experience, in this case, acts as one of
the essential goals of the development of the associated devices. However, regardless of the
improvement in the level of user experience through the introduction of voice-related
interactions, experts argue that such technological advancements in most cases are left and
assigned to the cloud providers.
The primary reason behind the assignment of these tasks and concepts to cloud providers
comes from the notion that they are complex to manage. The complexity of the idea of voice
recognition and interactions increase with the issues of security. The cloud providers may not
maintain the privacy of the users in the context. Since the cloud providers do not offer the
ANONYMIZING VOICE INPUT IN MOBILE DEVICES
9
required level of protection to the user data, experts believe that the introduction of voice
masking mechanisms, in the end, may create a conducive environment driven by the desire to
provide the ultimate level of privacy to the information while at the same time guaranteeing that
the identity of the communicating parties remains concealed.
Growth of voice interactions and sanitization
The growth of voice recognition and communications dates back years ago. The journey
of the use of voice recognition and interactions with the various devices has been long and
fruitful. In the past, people would not interact with their devices using voice messages and
commands. However, today, smart objects and people can communicate via voice commands
based on the adopted operating systems such as Android and features in the context.
Various pieces of research recognize the inherent abilities of the use of voice sanitizers to
improve the overall security of the affected populations while at the same time guaranteeing the
safety and privacy of nth excommunicating stakeholders. In addition to that, the continued
improvements in the field of technology and security bring about risks that may undermine the
privacy and the abilities of the voice sanitizers to perform and deliver their goals. As the paper
describes, the use of voice sanitizers, in the long run, must help the communicating parties to
enjoy privacy, which in the end, translates into improved security of data and identity.
In this context, the working of the selected models of voice sanitization follows various
strategies and techniques. For instance, looking at the example of voice-mask, one of the most
common tools used to sanitize communication between parties, it follows that this product uses
mechanisms that help to prevent the occurrence of various attacks that may happen affecting the
communication processes. In addition to that, the model acts as an intermediary between the
cloud and the users or communicating parties.
ANONYMIZING VOICE INPUT IN MOBILE DEVICES
10
For instance, the communication between two parties must go past through the model
before going to the cloud on either end. As an intermediary, the product makes it easy for the
communicating parties to guarantee the privacy of the members while eliminating risks that may
arise from the exposure of the information from the cloud providers. As highlighted above, cloud
providers usually exhibit weaknesses which can be used by attackers as the ultimate
vulnerabilities and avenues for accessing the communications carried out by the associated
people.
Potential benefits of voice sanitization
Based on the given information, one can confirm that the primary benefit that voice
sanitization offers to the users include improved privacy. The products selected in this context
must ensure that they conceal the identity of the communicating parties. The cloud platforms and
providers, at times, may compromise the communications and information provided by the users.
However, with the use of the voice sanitizers, it is possible to guarantee privacy while at the
same time improving the security of the communications carried out in the process. This paper
outlines some of the issues and concepts revolving around the implementation of voice sanitizers
based on the presented hypothesis that they play a critical role in improving the security and
privacy of the communicating parties.
The paper, therefore, analyzes the different strategies of protecting the speech data
privacy of voice input users without compromising their user experience. It also shows how this
particular research topic is SMART.
Problem Statement
The development of voice recognition and speech recognition strategies represents an
essential step in the human-computer interaction field. While technology has enabled the
ANONYMIZING VOICE INPUT IN MOBILE DEVICES
11
creation of algorithms for speech recognition, work is still necessary to ensure the safety of these
algorithms. In this sense, the use of voice input has come with significant privacy and security
risks for its users. The cloud or third parties responsible for obtaining voice data from the cloud
can link a user’s speech records to the actual individual by leveraging on the speech recognition
technology. If the cloud manages to collect several speech samples of a specific user from
various sources, e.g., YouTube, then train a voice model of the user, it may be able to identify
the user’s records from the speech database leading to an identity privacy breach.
Further, the cloud could also analyze the content of a user’s speeches to gather detailed
private information. A content privacy breach can get conducted if the cloud uses natural
language processing to collect data from a user’s voice command and voice search history or
even emails and text messages that were typed using voice input. Through this, the cloud can
accurately determine a users personal preferences, habits, demographic characteristics,
schedules, and communication. Security Risks could also arise such as spoofing attacks to
speech authentication systems because the use of voice for biometric information has risen in
modern authentication systems, e.g., unlocking smartphones, gaining access to apps, e.g.,
WeChat and activating services like OK Google or Hey Siri. Therefore, the cloud can repeatedly
collect a user’s voice samples to accurately create a voice print for synthesizing other speeches
identical to the user’s voice. Such can be used to spoof voice authentication systems and gain
unauthorized access to a user’s private information or even worse, be used to replicate the user’s
voice and create recordings for blackmailing the user. There is, therefore, need to come up with
mechanisms that can safeguard the privacy and the security of voice input users.
ANONYMIZING VOICE INPUT IN MOBILE DEVICES
12
Purpose Statement
As illustrated previously, the use of voice recognition and speech recognition require the
implementation of secure anonymizing features that minimize the risk imposed by these systems
to the safety and privacy of the individual. These anonymizing characteristics ensure that the
exact user cannot get identified through the analysis of the recorded voice, but only the content
of the recorded message. The research design used in the evaluation of the efficiency of the
anonymizing traits of voice and speech recognition system uses a qualitative approach focused
on the construction of grounded theory. The data collection method will thus involve the
completion of questionnaires by the …
Purchase answer to see full
attachment
