PERSONAL VOICE ASSISTANT

Sudhir Yadav
APEX Inst. Of Technology
Chandigarh University
Chandigarh,India
[email protected]

Abstract:-In this paper, Voice intelligent Assistance tool is and may not always provide a satisfying user
used for searching purposes, summary extraction, setting experience. Additionally, voice assistants collect
reminders just by using voice commands. Voice recognition large amounts of data, making them vulnerable to
technology allows us to access any document or file we cyberattacks. Personalization is also a challenge, as
desire. voice assistants may not always be able to
If the user spells out the word it automatically types in the personalize responses to individual users. Overall,
required field. It recognizes the speech and searches the personal voice assistants facenumerous challenges in
appropriate content in the database and retrieves it. The audio their early stages of development.
from the microphone will be collected by this voice assistant,
which will subsequently translate it into text. To ensure that II. LITERATURE SURVEY
the virtual assistant can comprehend them, the user should
choose the proper language. If any wrong or invalid Voice assistant has a long history with several waves
communication happens it invokes some messages in dialog of major innovations. Voice assistant for dictation,
box. It is a software application which performs tasks and search, and voice commands has become a standard
events based on commands. Voice-Command and speech feature on smartphones and wearable devices. The
synthesis are enhancing the level of user- interaction in study stems from an overlooking literature review in
applications. This intelligent personal assistant (IPA) can order to present generic knowledge (theory and
interact with the user by opening a report, providing a brief concepts) about voice control, virtual assistants,
summary via speech-to-text, and outlining the most crucial fields of use and more. When looking at a number of
details in the appropriate context. Here, an attempt is made to currently available intelligent programs with natural
build an intelligent voice personal assistant using Python, language processing capabilities, many examples can
which offers the ability to control voice- activated devices be found in everyday life filling a variety of roles.
and speech- activated smart devices for information The first speech recognition system, named Audrey,
extraction. was created by Bell Laboratories in 1952. Audrey
NLP (Natural Language Processing) helps the virtual was rather rudimental and limited technology wise,
assistant to understand and respond to human speech and understanding only ten digits - spoken by particular
based on the voice commands the tasks are performed. BERT people (Pieraccini, 2012). About 10 years later, IBM
is designed for the computers to understandthe meaning of developed and demonstrated their Shoebox Machine.
ambiguous language. Keywords:- Virtual Assistant, Speech The device recognized and responded to 16 different
Recognition, Extracti spoken words, including all ten digits “0” to “9” as
Keywords— Virtual Assistant, Speech Recognition, well as calculating commands such as “plus” or
Extractive summarization, BERT “minus” (IBM, 2018).Shoebox Machine recognized
and responded to 16 spoken words, including the ten
I. INTRODUCTION digits from “0” through “9”, only in English by a
designated speaker. These limitations later proved to
be problematic, increasing the scepticism opposing
The aim of this project is to develop a highly
voice recognition. Mid 1970’s came the Hidden
functional and user- friendly personal voice assistant
Markov Model (HMM) (Rabiner,1989). The HMM
that can effectively understand, process, and respond
considerably altered the development of a feasible
to user voice commands. The voice assistant should
speech recognition software. With the help of HMM
be capable of performing a wide range of tasks, from
speech recognition started using a statistical method
providing information and reminders to executing
measuring the probability of unknown sounds being
complex actions, all while maintaining a natural and
words. Now, the potential to recognize an unlimited
engaging interaction with the user. Personal voice
number of words became imminent due to the
assistants face challenges such as accuracy, limited
method allowing the number of understandable
functionality, privacy concerns, security risks, and
words go up to a few thousands. These choices of
user acceptance. They struggle with understanding
human speech, particularly in noisy environments,

XXX-X-XXXX-XXXX-X/XX/$XX.00 ©20XX IEEE

observation distribution in each state of the model possible with the assistance of algorithms designed to
allow accurate modelling of virtually unlimited types help the program learn from its inputs.
of data. The first mass accessible voice command
system was launched by Apple Inc. as they released
the virtual assistant named Siri in 2011 (Bostic,
REFRENCES
2013).The intelligent bot Siri can be found as
standard on Apple mobile devices now and is
considered a core component on these devices. Siri is [1] Harshit Agrawal, Nivedita Singh, Gaurav Kumar, Dr. Diwakar
Yagyasen, Mr. Surya Vikram Singh. "Voice Assistant Using Python" An
a personal assistant that uses natural language International Open Access-revied, Refereed Journal.Unique Paper ID:
processing to answer questions and outsource 152099, Publication Volume & Issue: Volume 8, Issue 2, Page(s): 419-
requests to web services that will then be carried out 423.
for the user. Similarly to this, the chatbot HAL was [2] George Terzopoulos, Maya Satratzemi “Voice Assistants and Smart
Speakers in Everyday Life and In Education”, Department of Applied
created by Zabaware Inc to function as a virtual Informatics, University of Macedonia, Thessaloniki, Greece.
assistant for users on computers. The bot also uses [3] Deepak Shende. Ria Umabiya, Monika Raghorte, Aishwarya Bhisikar.
natural language processing algorithms to converse Anup Bhange. "Al Based Voice Assistant Using Python", International
with the user and take notes from what the user is Journal of Emerging Technologies and Innovative Research
saying in an effort to organize the data given to it. (www.jetir.org), ISSN 2349-5162, Vol.6, Issue 2, page no.506-509,
February 2019.
IBM has invested a large amount of resources into
[4] Tulshan, Amrita & Dhage, Sudhir. (2019). “Survey on Virtual
this field and has created Watson, a system Assistant: Google Assistant, Siri, Cortana, Alexa”, 4th International
developed to compete on the TV show Jeopardy!. Symposium SIRS 2018, Bangalore, India, September 19–22, 2018,
This system exhibits the current capabilities of Revised Selected Papers. 10.1007/978-981-13- 5758-9_17.
intelligent systems with natural language recognition [5] Dr. Kshama V. Kulhalli, Dr.Kotrappa Sirbi, Mr. Abhijit J. Patankar,
"Personal Assistant with Voice Recognition Intelligence", International
as it successfully beat the two most successful Journal of Engineering Research and Technology. ISSN 0974-3154
human contestants of the show. In contrast to these Volume 10, Number 1 (2017”
roles is the chatbot Kari who functions as a virtual IEEE conference templates contain guidance text for
girlfriend. This system communicates with the user composing and formatting conference papers.
and through using similar methods natural language Please ensure that all template text is removed from
your
recognition tries to provoke social conversation with conference paper prior to submission to the
the user. conference. Failure to remove template text from
The software aims to give personal your paper may result in your paper not being
companionship and to replicate human interaction published.
as accurately as