lOMoARcPSD|40929137

CSDF MINI Project Report

Cyber Security And Digital Forensics (Savitribai Phule Pune University)

Scan to open on Studocu

Studocu is not sponsored or endorsed by any college or university

Downloaded by Rushikesh Hire ([email protected])
 lOMoARcPSD|40929137

A REPORT ON

Cyber Security and Digital Forensic

Mini Project

on

“Design and develop a tool for

digital forensic of Audio”

SUBMITTED TO THE SAVITRIBAI PHULE PUNE UNIVERSITY,

PUNE IN THE PARTIAL FULFILLMENT FOR THE AWARD OF

THE DEGREE OF BACHELOR OF ENGINEERING IN COMPUTER

ENGINEERING

BY

Tejas Ubale 4101077

Under the Guidance

Ms. G. N. Rukare

DEPARTMENT OF COMPUTER ENGINEERING

Sinhgad Institute of Technology & Science Narhe, Pune - 411041

Downloaded by Rushikesh Hire ([email protected])

 lOMoARcPSD|40929137

Department of Computer Engineering, Sinhgad Institute of Technology

and Science, Narhe

CERTIFICATE

This is to certify that the report entitled

“Design and develop a tool for digital forensic

of Audio”

Submitted by

Tejas Ubale 4101077

It is a bonafide work carried out by them under the supervision of Ms. G. N. Rukare and is
approved for the partial fulfillment of the requirement of the Laboratory Practices IV course in
Fourth Year Computer Engineering, in the academic year 2023-2024 prescribed by Savitribai
Phule Pune University, Pune.

Ms. G. N. Rukare Dr. G. S. Navale Dr. S. D. Markande

Guide Head of Department Principal

SINHGAD INSTITUTE OF TECHNOLOGY AND SCIENCE, NARHE, PUNE– 411041

Place: Pune
Date:

2
Downloaded by Rushikesh Hire ([email protected])
 lOMoARcPSD|40929137

Acknowledgement

I take this opportunity to acknowledge each and every one who contributed towards my
work. I express my sincere gratitude towards guide Ms. G. N. Rukare, Assistant
Professor at Sinhgad Institute of Technology and Science, Narhe, Pune, for her valuable
inputs, guidance and support throughout the course.
I wish to express my thanks to Dr. G. S. Navale, Head of Computer Engineering
Department, Sinhgad Institute of Technology and Science, Narhe for giving me all the
help and important suggestions all over the Work.
I thank all the teaching staff members, for their indispensable support and priceless
suggestions.
I also thank my friends and family for their help in collecting data without which their
help Database Management Laboratory report have not been completed. At the end my
special thanks to Dr. S. D. Markande, Principal Sinhgad Institute of Technology and
Science, Narhe for providing ambience in the college, which motivate us to work.

3
Downloaded by Rushikesh Hire ([email protected])
 lOMoARcPSD|40929137

Contents

SR Page
Topic Name
No. No.
1 Introduction 5
2 Problem Statement 7
3 Methodology 8
4 Working 10
5 Algorithm 11
6 Output 14
7 Conclusion 15
8 References 16

4
Downloaded by Rushikesh Hire ([email protected])
 lOMoARcPSD|40929137

1. Introduction

In today's digital age, the analysis and investigation of digital audio content play a
crucial role in various fields, including law enforcement, cybersecurity, and multimedia
content management. As the volume of audio data continues to grow exponentially, the
need for efficient and effective digital forensic tools for audio becomes increasingly
evident. The task of preserving, analyzing, and extracting meaningful information from
digital audio recordings is not only complex but also essential for legal proceedings,
security assessments, and content verification.

Designing and developing a specialized tool for digital forensic analysis of audio data is
a pressing need, as it enables forensic experts, investigators, and analysts to uncover
vital evidence, detect tampering or manipulation, and ensure the integrity and
authenticity of audio recordings. This tool will provide the means to streamline and
enhance the investigative process, making it more efficient, accurate, and
comprehensive.

This project aims to create a cutting-edge software solution that combines advanced
audio analysis techniques, data visualization, and user-friendly interfaces to empower
forensic experts and law enforcement agencies in their pursuit of truth and justice. By
leveraging state-of-the-art technologies, this tool will help in identifying audio
forgeries, recovering deleted content, verifying timestamps, and ensuring that audio
evidence complies with legal standards.

This endeavor represents a significant contribution to the evolving field of digital

forensics, addressing the unique challenges associated with audio data in a digital
context. Through the development of this tool, we intend to provide investigators with a
powerful resource to unravel the mysteries concealed within digital audio recordings,
ultimately serving the cause of justice, security, and the preservation of digital truth. In
the following sections, we will delve into the key features, design principles, and
technical aspects of this innovative digital audio forensic tool.

Downloaded by Rushikesh Hire ([email protected])

 lOMoARcPSD|40929137

Objectives:
The objectives of this project include:

● To design a user-friendly tool for digital forensic audio analysis.

● To implement algorithms for audio analysis, such as cost-benefit analysis, ethical
analysis, and test and validation.
● To create a reporting system to present findings effectively.
● To ensure the tool is versatile and can handle various audio.

Downloaded by Rushikesh Hire ([email protected])

 lOMoARcPSD|40929137

2. Problem Statement

The challenge at hand is to create an advanced digital forensic tool specifically designed for audio
analysis. This project addresses the growing demand for a specialized digital audio forensic tool. With
the rise of audio evidence in legal and cybersecurity contexts, current tools fall short. The goal is to
create an advanced, user-friendly tool tailored for precise audio analysis, ensuring data integrity and
improving the accuracy of examinations.

Downloaded by Rushikesh Hire ([email protected])

 lOMoARcPSD|40929137

3. Methodology

The methodology section outlines the step-by-step approach used in designing and developing the Digital
Forensic Audio Analysis Tool. Each sub-section describes a crucial aspect of the project's workflow:

Data Collection

In this stage, a diverse range of data was gathered to facilitate the development and testing of the tool. Data
sources included both publicly available audio content and proprietary datasets. This data encompassed
various audio formats, resolutions, and sources to ensure that the tool would be versatile and capable of
handling different types of audio evidence. The purpose of collecting this data was to provide the tool with
a wide spectrum of audio scenarios for comprehensive testing and analysis.

Data Processing

Designing and developing a tool for digital audio forensic data processing is a multifaceted endeavor focused on
creating a sophisticated software application capable of analyzing, preserving, and extracting valuable insights
from digital audio recordings. This tool encompasses a range of functionalities, from audio format conversion to
advanced voice analysis and tamper detection. It serves as a vital resource for forensic experts and investigators
in various fields, helping them ensure the authenticity and integrity of audio evidence, identify manipulation or
tampering, and uncover hidden information. In essence, it aims to streamline the complex process of audio data
analysis by incorporating advanced algorithms, user-friendly interfaces, and adherence to legal standards,
thereby enhancing the accuracy and efficiency of investigations related to digital audio recordings.

Downloaded by Rushikesh Hire ([email protected])

 lOMoARcPSD|40929137

Data Analysis

The core of the tool's functionality revolved around the application of audio analysis algorithms to the
preprocessed data. This involved the following key tasks:

Tampering Detection: Algorithms were implemented to identify signs of tampering, such as cut-and-paste
operations, inconsistencies in the audio timeline, or alterations in visual elements. The goal was to detect any
potential audio manipulation accurately.

Metadata Analysis: The extracted metadata was analyzed to gather information about the video's history,
origin, and editing. This step helped in establishing the context and authenticity of the audio content.

Evidence Identification: The tool was designed to identify potential evidence within the audio. This could
include objects, faces, or other elements of interest. The ability to automatically identify evidence could
significantly expedite investigations.

Reporting

Designing and developing a tool for digital audio forensic reporting involves creating a software solution
capable of systematically organizing, summarizing, and presenting the findings of audio forensic analyses.
This tool streamlines the process of creating comprehensive reports, making it essential for forensic experts
and investigators. It ensures that crucial audio evidence is meticulously documented and presented in a clear
and legally admissible manner, thereby enhancing the credibility and effectiveness of the investigative
process.

Downloaded by Rushikesh Hire ([email protected])

 lOMoARcPSD|40929137

4. Working

1. Audio Input: The code starts by specifying the path to an audio file, which is "church.mp4" in this case.

2. Audio Capture: It uses the OpenCV library to open and read the audio file. The
`cv2.VideoCapture` function is used to initialize the audio capture object.

3. Frame Collection:
- It creates an empty list `frames` to store the frames of the video.
- A `while` loop reads frames from the audio one by one using `cap.read()`.
- The loop continues until there are no more frames (`ret` becomes `False`), and each frame
is appended to the `frames` list.

4. Release Video: After all frames are collected, the audio capture object is released using
`cap.release()` to free up system resources.

5. Frame Processing:
- For each frame in the `frames` list, the code performs the following steps:
- It converts the frame to grayscale using `cv2.cvtColor` for simplifying the frame for comparison.
- The grayscale frame is resized to a small 8x8-pixel resolution.
- The average pixel value of the resized frame is calculated.
- A binary hash value is created based on whether each pixel in the resized frame is greater than or
equal to the average pixel value. This results in a binary hash where 1 represents a pixel value
greater than or equal to the average, and 0 represents a pixel value less than the average.
- The binary hash is appended to the `hashes` list for each frame.

6. Duplicate Frame Detection:

- The code creates an empty set called `duplicate_frames` to store the indices of duplicate frames.
- It uses nested loops to compare each pair of frames by their hash values.
- If the hash values of two frames are equal, meaning they have the same binary hash, their indices
are added to the `duplicate_frames` set.

7. Display Duplicate Frames:

- The code iterates through the indices of duplicate frames in the `duplicate_frames` set.
- It uses Matplotlib to display the corresponding frames as images.
- The title of each displayed image indicates that it's a duplicate frame, and the frame's index is shown.
- The `plt.axis('off')` command turns off axis labels and ticks in the displayed image.

10

Downloaded by Rushikesh Hire ([email protected])

 lOMoARcPSD|40929137

5. Algorithm

import wave

import audioop

def vad(audio_file):

# Read the audio file

audio = wave.open(audio_file, 'rb')

frame_rate = audio.getframerate()

sample_width = audio.getsampwidth()

# Initialize variables

buffer_size = 1024

speech_segments = []

start_time = None

end_time = None

while True:

frames = audio.readframes(buffer_size)

if not frames:

break

11

Downloaded by Rushikesh Hire ([email protected])

 lOMoARcPSD|40929137

# Calculate energy of the frame

energy = audioop.rms(frames, sample_width)

if energy > 500: # Adjust threshold as needed

if start_time is None:

start_time = audio.tell() / frame_rate

else:

if start_time is not None:

end_time = audio.tell() / frame_rate

speech_segments.append((start_time, end_time))

start_time = end_time = None

# Close audio file

audio.close()

return speech_segments

# Example usage

if __name__ == "__main__":

audio_file = 'sample.wav'

detected_segments = vad(audio_file)

12

Downloaded by Rushikesh Hire ([email protected])

 lOMoARcPSD|40929137

if detected_segments:

print("Detected Speech Segments:")

for i, (start, end) in enumerate(detected_segments, start=1):

print(f"Segment {i}: {start:.1f}s - {end:.1f}s")

else:

print("No speech detected.")

13

Downloaded by Rushikesh Hire ([email protected])

 lOMoARcPSD|40929137

6. Output

Fig.1 Speech Segments Detected

Fig.2 No Speech Detected

14

Downloaded by Rushikesh Hire ([email protected])

 lOMoARcPSD|40929137

CONCLUSION

In this project, we successfully developed a Python-based solution for detecting duplicate frames in
audiocontent. Leveraging the power of OpenCV for audioprocessing and NumPy for numerical
operations, we first collected and processed audioframes. The key innovation was the creation of
binary hash values for each frame, simplifying the frames for efficient comparison.

In conclusion, the design and development of a specialized tool for digital audio forensic analysis
represents a significant advancement in the field of digital forensics. Such a tool addresses the
increasing demand for efficient, accurate, and comprehensive audio evidence examination, crucial in
legal proceedings, cybersecurity investigations, and content verification. By incorporating advanced
algorithms, user-friendly interfaces, and adherence to legal standards, this tool empowers forensic
experts and investigators to ensure the authenticity and integrity of audio recordings, detect tampering,
and extract hidden information. It not only streamlines the investigative process but also enhances the
accuracy and efficiency of audio data analysis. The development of this tool exemplifies the synergy
between cutting-edge technology and the pursuit of truth, justice, and security in the digital age,
marking a significant contribution to the evolving landscape of digital forensic practices.

15

Downloaded by Rushikesh Hire ([email protected])

 lOMoARcPSD|40929137

REFERENCES

[1] https://www.anaconda.com/
[2] https://jupyter.org/
[3] https://www.geeksforgeeks.org/
[4] https://www.w3schools.com/

16

Downloaded by Rushikesh Hire ([email protected])