A SYNOPSIS ON

SPEECH-TO-TEXT AND VOICE ACTIVATED INTERFACE

Submitted in partial fulfilment of the requirement for the award of the degree of

Master OF COMPUTER APPLICATION

Submitted by:

Name: Ankit Chauhan University Roll No.: 1102994

Under the Guidance of

Miss. Gunjan Mehra
Ass Professor

Department of Computer Science and Engineering

Graphic Era (Deemed to be University) Dehradun,
Uttarakhand May-2025

1
 CANDIDATE’S DECLARATION
I hereby certify that the work which is being presented in the Synopsis entitled “SPEECH-
TOTEXT AND VOICE ACTIVATED INTERFACE” in partial fulfilment of the
requirements for the award of the Degree of Masters Of Computer Application in the
Department of Computer Application of the Graphic Era (Deemed to be University), Dehradun
shall be carried out by the undersigned under the supervision of Miss. Gunjan Mehra,
Assistant Professor, Department of Computer Application, Graphic Era (Deemed to be
University), Dehradun.

Ankit Chauhan 1102994 Signature

The above-mentioned students shall be working under the supervision of the undersigned on
the “SPEECH-TO-TEXT AND VOICE ACTIVATED INTERFACE”

Signature Signature
Supervisor Head of the Department

Internal Evaluation (By DPRC Committee)

Status of the Synopsis: Accepted / Rejected

Any Comments:

Name of the Committee Members: Signature with Date

1.
2.

2
 Table of Contents

Chapter No. Description Page No.

Chapter 1 Introduction and Problem Statement 4

Chapter 2 Background/Literature Survey 5

Chapter 3 Objectives 6

Chapter 4 Hardware and Software Requirements 7

Chapter 5 Algorithms 8

References 9

3
Chapter 1 Introduction

In the following sections, a brief introduction and the problem for the work has been included.

1.1 Introduction
The SPEECH-TO-TEXT AND VOICE ACTIVATED INTERFACE is an
intelligent desktop application that interprets spoken language and responds
accordingly using a graphical interface. This project combines speech
recognition, natural language processing, and machine learning to allow users
to interact with the system through voice commands.

This assistant is designed to be user-friendly and helpful in daily tasks such as

searching Wikipedia, playing music, opening websites like YouTube, or simply
converting speech into text. The assistant responds with both visual feedback via
a GUI and audible speech via text-to-speech synthesis.

1.2 Problem Statement

In today’s digital age, human-computer interaction is moving increasingly toward
natural interfaces such as voice. The ability to control devices and access information
hands-free not only enhances convenience but is also critical in scenarios where
manual input is difficult, such as while multitasking or for users with physical
impairments.
However, many existing voice assistant solutions (e.g., Alexa, Google Assistant) are
heavily cloud-dependent, raise privacy concerns, and are not easily customizable for
specific user needs or offline use.
This project aims to develop a simple, customizable, offline-friendly speech-enabled
virtual assistant that can interpret user speech, understand the user’s intent using
machine learning, and carry out basic tasks such as searching Wikipedia, playing
music, or opening websites — all through a user-friendly GUI.

4
Chapter 2

Background/Literature Survey
Voice assistants have become an integral part of human-computer interaction, with popular examples
such as Google Assistant, Amazon Alexa, Apple Siri, and Microsoft Cortana. These systems
utilize natural language processing (NLP), machine learning (ML), and voice synthesis to
understand and respond to user queries. While highly effective, they are often cloud-dependent and
integrated into specific ecosystems, making them less customizable for individual users or desktop
applications.

3.1 Existing Systems and Their Limitations

• Google Assistant / Alexa / Siri: These assistants rely on internet connectivity and largescale
cloud-based models. They are optimized for mobile and IoT devices, limiting their scope for
offline or desktop-specific customizations.

3.2 Key Technologies Studied

• Speech Recognition: Libraries like speech_recognition (Google API, CMU Sphinx) convert
spoken words into text. Accuracy depends on ambient noise and model quality.
• Text-to-Speech (TTS): Tools like pyttsx3 allow voice synthesis using the local SAPI5
engine, supporting customizable voices and offline operation.
• Machine Learning for NLP: ML models such as Logistic Regression, Naïve Bayes, and
SVM can classify user intents based on training data using TF-IDF or CountVectorizer
techniques.
• GUI Integration: Tkinter enables interactive GUI development, which is essential for
desktop applications needing visual feedback beyond voice.

5
Chapter 3

Objectives
The primary objective of this project is to develop an interactive voice-controlled assistant
application that can recognize and interpret spoken commands, provide information retrieval through
Wikipedia searches, and assist with media playback through YouTube. This application aims to
leverage natural language processing and machine learning techniques to classify user intents from
speech input, thereby offering a hands-free, intuitive user experience. The assistant is designed with
a user-friendly graphical interface using tkinter, enabling seamless interaction.
Key Objectives:
1. Speech Recognition Integration:
o Implement accurate real-time speech-to-text conversion using the
speech_recognition library and Google's speech API.
o Ensure the assistant can handle unclear or ambiguous audio inputs gracefully with
proper feedback.
2. Intent Classification:
o Utilize a pre-trained machine learning model to analyze the transcribed user
commands.
o Predict the user’s intent (such as searching Wikipedia, playing music, opening
YouTube, or exiting the app) based on natural language input.
o Employ a vectorizer for text preprocessing to enhance model prediction accuracy.
3. Wikipedia Information Retrieval:
o Automatically extract relevant search topics from user commands. o Retrieve
and summarize concise Wikipedia articles on the requested topics.
o Handle disambiguation cases and missing pages with appropriate user notifications.
4. Media Playback Automation:
o Facilitate music or video search on YouTube by voice command.
o Automatically open YouTube in the web browser and initiate playback using
keyboard automation (pyautogui).
5. Text-to-Speech Feedback:
o Provide spoken responses and confirmations to enhance user interaction and
accessibility using the pyttsx3 library.
o Maintain an adjustable speech rate for clarity.
6. User Interface Design:
o Create an intuitive and visually appealing GUI using tkinter with components for
displaying recognized speech, interaction status, and buttons for recording and
exiting.

6
 o Prevent GUI freezing by using threading to manage blocking operations such as
listening and processing.

Chapter 4

Hardware and Software Requirements

4.1 Hardware Requirements

Sl. No Name of the Hardware Specification

1 Processor Intel Core i3 or higher

2 RAM Minimum 8 GB

3 Microphone & Speakers Built-in or external for voice input/output

4.2 Software Requirements

Sl. No Name of Software Specification

1 Python Version 3.8 or above

2 Python Library tkinter , speech_recognition , threading , wikipedia , pyttsx3,

webbrowser ,pyautogui, joblib , Wikipedia

3 Operating System Window 10 or Higher

7
Chapter 5

Algorithm

Step 1: Initialize System

• Load ML model and vectorizer (used for classifying user commands).
• Initialize Text-to-Speech (TTS) engine with desired voice settings.
• Set up GUI with input/output display and control buttons.

Step 2: Greet the User

• Check the current time (morning, afternoon, evening).
• Use TTS to greet the user accordingly.

Step 3: Listen to User Command

• Activate microphone.
• Use speech_recognition library to capture and convert voice to text.

Step 4: Predict User Intent

• Pass the user command text to the ML model.
• Use the trained model to predict the intent (e.g., search Wikipedia, play music, send email).

Step 5: Perform the Appropriate Action Based

on the predicted intent:
• Wikipedia → Search and read summary.
• Play Music → Open YouTube and search.
• Exit → Say goodbye and close the application.
• Normal Text Conversion → Print the text.

Step 6: Display Output •

Show the assistant’s responses in the GUI text area.
• Use TTS to read out the response.

Step 7: Loop
• Continue listening and responding until user gives exit command.

8
Reference
1. Google Cloud Speech-to-Text API Documentation https://cloud.google.com/speech-to-
text
[Useful reference for understanding cloud-based voice processing.]

2. Python SpeechRecognition Library Docs https://pypi.org/project/SpeechRecognition/

[Library used for capturing and processing voice input.]

3. Python Tkinter Library https://docs.python.org/3/library/tkinter.html [For GUI

development references.]

4. W3Schools – Machine Learning with Python

https://www.w3schools.com/python/python_ml_getting_started.asp
[Beginner-friendly guide to ML concepts.]

5. Chatgpt [For error correction and verification]