Volume 7, Issue 2, February – 2022 International Journal of Innovative Science and Research Technology

ISSN No:-2456-2165

Library Management Using Voice Assistant

Samiksha Gaikwad Shriya Purandare
Department of Computer Engineering Department of Computer Engineering
International Institute of Information Technology International Institute of Information Technology
Pune, India Pune, India

Swathi Balaji Kimi Ramteke

Department of Computer Engineering Department of Computer Engineering
International Institute of Information Technology International Institute of Information Technology
Pune, India Pune, India

Abstract:- A Personal Assistant is a computer application to human speech, (B) comprehend what is being indicated, and
that uses Artificial Intelligence (AI) to assist humans with (C) conduct an action or respond with their own synthesized
their tasks. In today's world, the use of a voice personal voicection. [4].
assistant is becoming more common. Modern IPAs are
capable of a broad range of functions, from simple ones like Natural Language Processing (NLP) is mostly about
opening an app or setting an alarm to more complicated instructing machines to understand human languages and derive
ones like taking notes. meaning from texts at their most basic level. Text mining, text
Siri from Apple, Google Assistant and Alexa from Amazon categorization, text and emotion monitoring, and speech
are all examples of AI assistants. production and identification are just a few of its many
applications. This is also why Natural language processing We
Keywords:- Intelligent Personal Assistant, NLP, ASR, Location constructed a working voice assistant for the purpose of this
of Books, Library Management. article, which can do simple tasks such as locating a book in the
library. It can interpret "audio instructions" and obtain
I. INTRODUCTION information from a database.

A student's life revolves around the library. It is also II. LITERATURE SURVEY
beneficial to academic researchers. Libraries may now be found
in almost any location. It's a huge endeavor to locate a book at In comparison to previous assistants, he implemented a lot
the library. While looking for the needed book, some people of stuff. It is quite beneficial in human life nowadays. It's a
lose interest. It takes a long time to look for books on the really straightforward application. It is also used in the
computer. corporate world, for example, in laboratories where people
wear gloves and bodysuits for safety reasons, making it
With open source software products like Moodle for impossible to write. However, with a voice assistant, they can
Virtual Learning gaining traction in related fields, many access whatever information they need, making their work
librarians are looking for OSS alternatives to their present easier. Only the most fundamental elements have been
Library Management Systems. implemented in this study. For example, a Google term search,
a YouTube song/video, a location search, and current news.
An Intelligent Personal Assistant (IPA) is a computer There is a lot more that can be done. [1]
application that uses Artificial Intelligence (AI) to help people
do tasks. The IPA maintains a continuous dialogue with its The focus of the research is on the technique used to create
users while responding to their questions or carrying out a multilingual and adjustable voice recognition and speech
measures to fulfill their demands [2]. Modern IPAs are capable synthesis system. There can be no assumptions about the
of a broad range of functions, from simple ones like opening an language identification of vocabulary items in voice calling,
app or setting an alarm to more complicated ones like taking although the phone book can contain names in many languages.
notes or making phone calls. Google Assistant from Google, Each voice tag must be trained separately by the user. This
Siri from Apple, and Alexa from Amazon are all instances of keeps the amount of voice tags to a bare minimum. few, and the
IPAs [3]. person has trouble recalling the exact statement that was uttered
during the course of the training. [3]
Although IPAs are not required in terms of communicating
exclusively through voice, many current IPAs are pursuing With the desire for human-machine interactions, current
Voice User Interface, which involves engaging with users only voice recognition applications are becoming increasingly
through voice, without the necessity of displays or physical widespread. On traditional general-purpose computers, several
interaction [3]. This necessitates the IPA's ability to (A) listen speech-based interactive software programmes were run. It is

IJISRT22FEB186 www.ijisrt.com 236

Volume 7, Issue 2, February – 2022 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
speaker-dependent and only useful to a select group of people To make the system status clearer, these gadgets should
for whom the system has been trained. This type of technology deliver more effective multimodal feedback. [16]
provides good performance for a known user, but performance
drops exponentially for new users. [11] The interactions of well known voice assistants with 26
children are described in this study. These kids were able to
The design aspects of a speech recognition engine converse and play with the assistance.
configured for mobile devices are outlined in this study. New
features and crucial enhancements of the original concepts, Later on, the kids provide comments about the assistant.
which were ignored, are now discussed, despite the fact that Overall, all of the youngsters were enthusiastic helpers. We
main elements have previously been provided. We also found that the vast majority of participants had difficulty getting
demonstrate how these strategies may be successfully coupled the agents to grasp their inquiries. Several students attempted to
to fulfill a variety of design goals with minimal influence on raise their voice volume or add additional pauses to their
recognition performance. Speaker adaptation is a required remarks in the hopes of making their questions more
process for optimizing recognition performance. understandable. However, it does not always result in greater
This set is utilized to compute the replacements for the agent recognition. [17]
original parameters once enough adaptation utterances have
been observed. III. PROPOSED WORK
However, doing so will significantly increase the amount
of memory required for storing acoustic models. [12] We employ speech recognition to discover books in our
project, which makes this mammoth process a lot simpler. Even
This article discusses recent advancements in voice those with no prior understanding of the system can use it to
recognition technology as well as the author's opinions on the discover books more quickly. It cuts down on time spent
subject. Dictation and human-computer conversation systems looking for books and makes the library more user-friendly.
are the two primary domains where voice recognition
technology is used. Automatic broadcast news transcription is The system will connect with its users in real time as it
now being researched in the dictation area, particularly under responds to their questions. Inquiries include book locations
the DARPA project. The most pressing issue is determining based on a unique identifier or the title of the book.
how to make speech recognition systems resistant to acoustic
and linguistic diversity in speech. In this setting, a paradigm
change from speech recognition to understanding where the
speaker's underlying messages, that is, the meaning/context that
the speaker intended to express, are extracted rather than all the
spoken words, will be critical. [13]

By giving a short survey on Fully automated Speech

Recognition and discussing the major themes and progress
made in the past 60 years of research, this paper provides a rapid
technology perspective and a gratitude of the profound work
that has been done in this important area of speech
communication. Speech recognition has piqued the interest of
scientists as a vital field. It has had a technical influence on
society, and is projected to grow in this field of human-machine Fig. 1. Architecture Diagram.
interaction in the future. [14]
The steps taken will be as follows :
This article provides an overview of the various types of Step 1 : User Speaks
voice assistants as well as the functions that voice assistants User speaks into the microphone and then these signals are then
may perform. Before voice assistants can be used for anything sent to the processor.
Step 2 : Audio Processing (ASR and NLP)
that requires secrecy, privacy and security safeguards will need
to be strengthened. [15] Here, Automatic Speech Recognition (ASR) is used which
takes an audio file and converts it into a sequence of words. The
A collection of 17 criteria for voice-based devices is NLP module picks up keywords and identifies the book and its
described in this research. The heuristics cover a large portion location in the database.
of the design space and may also be used to generate early Step 3 : Relevant information gathered from the database
concepts. Error handling should receive more attention. Information is retrieved from the database (book and its
location).
Step 4 : Response/Error is received
The response is received through the speaker.

IJISRT22FEB186 www.ijisrt.com 237

Volume 7, Issue 2, February – 2022 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
IV. MODULES E. Pyttsx3
In a software, this is used to convert text to speech.
A. Automatic Speech Recognition (ASR) module
Speech recognition is defined as the automatic recognition F. Datetime
of human speech and is recognized as one of the most important Used to know date and time. It comes built in with Python.
tasks when it comes to making applications based on Voice
User Interfacing (VUI). G. Speech Recognition
Used to make the assistant understand your voice.
Python comes with several libraries which support speech
recognition. We will be using the speech recognition library V. USES AND APPLICATION
because it is the simplest and easiest to learn.
This voice based assistant will be very useful for a student
This module involves 2 steps - to search a book in the library.Voice based assistant will
communicate with the user and will give the exact location of
 Loading Audio the book at the library. This assistant can be built in various
To load audio the AudioFile function is used. The function colleges to reduce the work performed by the librarian.
opens the file, reads its contents and stores all the information
in an AudioFile instance called source. VI. SCOPE

We will traverse through the source and do the following In the near future, we can implement a system to keep the
things: records of book transactions. For example, the date of a book
 Every audio has some noise involved which can be removed borrowed and returned can be recorded. The voice assistance
using the adjust_for_ambient_noise function. can be made multilingual so that the user can give input in any
 Making use of the record method which reads the audio file of the languages. We can also implement a student and faculty
and stores certain information into a variable to be read later in- out system which will keep the record of entry and exit time
on. in the library.

 Reading data from audio VII. CONCLUSION

We can now invoke recognize_google() method and
recognize any speech in the audio. After processing the method The method described is appropriate for developing an
returns the best possible speech that the program was able to Intelligent Voice Assistant that can interpret domain-specific
recognize from the first 100 seconds. The output comes out to direct voice commands from users. We will be using this
be a bunch of sentences from the audio which turn out to be approach to build an automated library management system
pretty good. The accuracy can be increased by the use of more based on an intelligent voice assistant.
functions but for now it does the basic functionalities.
REFERENCES
B. Natural Language Processing (NLP) module
SpaCy is a Natural Language Processing package for [1]. Subhash S, Prajwal N Srivatsa, Siddesh S, Ullas A,
Python that is open-source. It’s primarily intended for use in the Santhosh B, "Artificial Intelligence- based Voice
workplace, where it may create real-world initiatives and Assistant" 2020, IEEE.
manage massive amounts of text data. Moreover, sixty [2]. J. Kurpansky, "What Is an Intelligent Digital Assistant?"
languages are supported by spaCy, which has prepared pipelines Medium, 2017.
for various languages and activities. Because this toolset is [3]. J. Iso-Sipila, M. Moberg and O. Viikki, "Multi-lingual
designed in Python and Cython, it is quicker and more accurate speaker independent voice user interface for mobile
when dealing with massive amounts of text data. devices," in ICASSP 2006. IEEE, 2006, vol. I, pp. 1081-
1084.
C. Subprocess [4]. “Intelligent Personal Assistants. In: Speech
Utilized to obtain information on system subprocesses, Technology" S. Springenberg, Seminar, Institute of
which are then used in different commands like shutdown, Computer Science Hamburg, 2016, pp. 5-7
sleep, and so on. Python includes this module by default. [5]. Saadman Shahid Chowdhury,Atiar Talukdar,
Ashik Mahmud, Tanzilur Rahman, "Domain Specific
D. WolframAlpha Intelligent Personal Assistant with Bilingual Voice
Wolfram's algorithms, knowledgebase, and AI Command Processing" TENCON 2018 - 2018 IEEE
technologies are used to compute expert-level responses. Region 10 Conference

IJISRT22FEB186 www.ijisrt.com 238

Volume 7, Issue 2, February – 2022 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
[6]. Ishita Garg Hritik Solanki Sushma Verma, [18]. Thimmaraja Yadava, Pravenn Kumar, H S Jayanna, "An
"Automation and Presentation of Word Document Using End to End Spoken Dialogue System to Access the
Speech Recognition" 2020 International Conference for Agricultural Commodity Price Information in Kannada
Emerging Technology (INCET) [7]Dinesh K, Prakash R Language/Dialects" [20]Ekenta Elizabeth Odokuma,
and Madhan Mohan P, "Real-time Multi Source Speech Orluchukwu Great Ndidi, "Development of a Voice-
Enhancement for Voice Personal Assistant by using Linear Controlled Personal Assistant for the Elderly and
Array Microphone based on Spatial Signal Processing" Disabled" IRJCS, 2016 [21]Vassil Vassilev, Anthony
2019 International Conference on Communication and Phipps, Matthew Lane, Khalid Mohamed, Artur
Signal Processing (ICCSP) Naciscionis "Two-factor authentication for voice
[7]. Alisha Pradhan and Amanda Lazar, University of assistance in digital banking using public cloud services"
Maryland Leah Findlater, University of Washington, "Use April, 2020
of Intelligent Voice Assistants by Older Adults With Low [19]. Shinya Iizuka, Kosuke Tsujino, Shin Oguri, Hirotaka
Technology Used" ACM Transactions on Computer- Furukawa, "Speech Recognition Technology and
Human Interaction Applications for Improving Terminal Functionality and
[8]. George Terzopoulos Maya Satratzemi, "Voice Assistant Service Usability" 2012 NTT DOCOMO Technical
and Artificial Intelligence in Education" Journal Vol. 13 No. 4
[9]. The 9th Balkan Conference Juha Iso-Sipilä, Marko [20]. Alisha Pradhan1, Kanika Mehta1, Leah Findlater2,
Moberg, Olli Viikki, "Multilingual Speaker-Independent "Accessibility came by accident:Use of voice controlled
Voice User Interface for Mobile Devices" Nokia intelligent personal assistants by people with disabilities"
Technology Platforms, Tampere, Finland April 21–26, 2018, Montréal, QC, Canada
[10]. D Satya Ganesh, Dr. Prasant Kumar Sahu School of [21]. Hasrat Jahan, "Reshaping learning through Artificial
Electrical Sciences Indian Institute of Technology Intelligence" Department of Education, Bhopal Degree
Bhubaneswar, Odisha, India, "A Study on Automatic College, 393, Ashok Vihar, Bhopal - 462 023 (M.P.) India.
Speech Recognition Toolkits" 2015 IEEE at
ICMOCE
[11]. Marcel Vasilache, Juha Iso-Sipilä, Olli Viikki, "On a
Practical Design of a Low Complexity Speech
Recognition Engine" 2014 IEEE at ICASSP
[12]. Sadaoki Furui Department of Computer Science Tokyo
institute of Technology, "Automatic Speech Recognition
and its Application to Information Extraction" 2-12-1,
Ookayama, Meguro-ku, Tokyo, 152-8552 Japan
[13]. M.A.Anusuya, S.K.Katti Department of Computer
Science and Engineering Sri Jayachamarajendra College
of Engineering, Mysore, India, "Speech Recognition by
Machine : A Review" IJCSIS Vol. 6, No. 3, 2009
[14]. Matthew B. Hoy, "Alexa, Siri, Cortana, and More: An
Introduction to Voice Assistants" Medical Reference
Services Quarterly, 37:1, 81-88, DOI:
10.1080/02763869.2018.1404391
[15]. Zhuxiaona Wei, James A. Landay, "Evaluating Speech-
Based Smart Devices Using New Usability Heuristics"
Published by the IEEE Computer Society April–June
2018
[16]. Stefania Druga, Cynthia Breazeal, Randi Williams,
Mitchel Resnick, "‘Hey Google is it OK if I eat you?’
Initial Explorations in Child-Agent Interaction" IDC
2017, June 27–30, 2017, Stanford, CA, USA
[17]. Polyakov E.V., Mazhanov M.S., Rolich A.Y., Voskov L.S.,
Kachalova M.V., Polyakov S.V., "Investigation and
Development of the Intelligent Voice Assistant for the
Internet of Things Using Machine Learning" 2018 IEEE

IJISRT22FEB186 www.ijisrt.com 239