7/22/24, 2:51 PM DSPA_ET22BTEC046_LAB3.

ipynb - Colab

pip install pydub ffmpeg-python

Collecting pydub
Downloading pydub-0.25.1-py2.py3-none-any.whl (32 kB)
Collecting ffmpeg-python
Downloading ffmpeg_python-0.2.0-py3-none-any.whl (25 kB)
Requirement already satisfied: future in /usr/local/lib/python3.10/dist-packages (from ffmpeg-python) (0.18.3)
Installing collected packages: pydub, ffmpeg-python
Successfully installed ffmpeg-python-0.2.0 pydub-0.25.1

(a) Load a short audio file using the inbuilt function in Python.

import IPython.display as ipd

from pydub import AudioSegment

audio_file = "/music.wav"

# Load audio file

audio = AudioSegment.from_file(audio_file)

# Play audio using IPython.display

ipd.Audio(data=audio_file, rate=audio.frame_rate)

0:01 / 1:22

from pydub import AudioSegment

import numpy as np
import matplotlib.pyplot as plt
audio_file = "/fantastic.wav"
audio = AudioSegment.from_file(audio_file)
duration_ms = len(audio)
print(f"Duration of the audio: {duration_ms} milliseconds")

samples = audio.get_array_of_samples()

# Calculate duration (in seconds) and sample rate

duration = len(samples) / audio.frame_rate
sample_rate = audio.frame_rate

# Create a time vector spaced linearly with the duration of the audio
time = np.linspace(0, duration, num=len(samples))

# Plotting the waveform

plt.figure(figsize=(10, 4))
plt.plot(time, samples, color='b')
plt.xlabel('Time (s)')
plt.ylabel('Amplitude')
plt.title('Audio Waveform')
plt.grid(True)
plt.show()

https://colab.research.google.com/drive/10hNM-QiWZCSknm9v49nXRXHuJaHWEML4?authuser=3#scrollTo=la4RB8soOwhM&printMode=true 1/7
7/22/24, 2:51 PM DSPA_ET22BTEC046_LAB3.ipynb - Colab

Duration of the audio: 4339 milliseconds

(b) Plot the time domain waveform with amplitude as a function of no of samples.

from pydub import AudioSegment

import numpy as np
import matplotlib.pyplot as plt

# List of audio files

audio_files = [
"/fantastic.wav",
"/music.wav",
"/oboe.wav"
]

# Initialize a list to store durations and samples of each audio file

durations = []
samples_list = []

# Loop through each audio file

for audio_file in audio_files:
# Load audio file
audio = AudioSegment.from_file(audio_file)

# Calculate duration (in seconds) and sample rate

duration = len(audio) / 1000 # Convert duration to seconds
sample_rate = audio.frame_rate

# Extract raw audio data (as numpy array)

samples = np.array(audio.get_array_of_samples())

# Store duration and samples

durations.append(duration)
samples_list.append(samples)

# Plotting the waveform for each audio signal

plt.figure(figsize=(15, 10))

for i in range(len(audio_files)):
plt.subplot(len(audio_files), 1, i+1)
plt.plot(np.linspace(0, durations[i], num=len(samples_list[i])), samples_list[i], color='b')
plt.xlabel('Time (s)')
plt.ylabel('Amplitude')
plt.title(f'Audio Signal {i+1} Waveform')
plt.grid(True)

plt.tight_layout()
plt.show()

https://colab.research.google.com/drive/10hNM-QiWZCSknm9v49nXRXHuJaHWEML4?authuser=3#scrollTo=la4RB8soOwhM&printMode=true 2/7
7/22/24, 2:51 PM DSPA_ET22BTEC046_LAB3.ipynb - Colab

(c) Plot the corresponding spectrogram of the signal

https://colab.research.google.com/drive/10hNM-QiWZCSknm9v49nXRXHuJaHWEML4?authuser=3#scrollTo=la4RB8soOwhM&printMode=true 3/7
7/22/24, 2:51 PM DSPA_ET22BTEC046_LAB3.ipynb - Colab
from pydub import AudioSegment
import numpy as np
import matplotlib.pyplot as plt
from scipy import signal

# List of audio files

audio_files = [
"/fantastic.wav",
"/oboe.wav",
"/music.wav"
]

# Initialize lists to store durations and samples of each audio file

durations = []
samples_list = []

# Loop through each audio file

for audio_file in audio_files:
# Load audio file
audio = AudioSegment.from_file(audio_file)

# Calculate duration (in seconds) and sample rate

duration = len(audio) / 1000 # Convert duration to seconds
sample_rate = audio.frame_rate

# Extract raw audio data (as numpy array)

samples = np.array(audio.get_array_of_samples())

# Store duration and samples

durations.append(duration)
samples_list.append(samples)

# Plotting the spectrogram for each audio signal

plt.figure(figsize=(10, 6))

for i in range(len(audio_files)):
# Compute spectrogram using scipy.signal.spectrogram
f, t, Sxx = signal.spectrogram(samples_list[i], fs=sample_rate, nperseg=1024, noverlap=512)

# Plot spectrogram
plt.subplot(len(audio_files), 1, i+1)
plt.pcolormesh(t, f, 10 * np.log10(Sxx), shading='gouraud') # Convert to dB and use gouraud shading
plt.colorbar(label='Intensity (dB)')
plt.ylabel('Frequency [Hz]')
plt.xlabel('Time [s]')
plt.title(f'Spectrogram of Audio Signal {i+1}')
plt.ylim(0, 10000) # Limit y-axis to 0-10kHz for better visualization
plt.tight_layout()

plt.show()

https://colab.research.google.com/drive/10hNM-QiWZCSknm9v49nXRXHuJaHWEML4?authuser=3#scrollTo=la4RB8soOwhM&printMode=true 4/7
7/22/24, 2:51 PM DSPA_ET22BTEC046_LAB3.ipynb - Colab

(d) Analyse the waveform to obtain the signal duration, sampling rate and number of samples.

import IPython.display as ipd

from pydub import AudioSegment

def play_audio(audio_files):
# Print options for the user
print("Choose an audio file to play:")
for i, file in enumerate(audio_files):
print(f"{i+1}: {file}")

# Prompt user to choose an audio file

choice = int(input("Enter your choice (1-3): ")) - 1 # Subtract 1 to match index

if choice < 0 or choice >= len(audio_files):

print("Invalid choice. Please choose a number between 1 and 3.")
return

# Load selected audio file

audio_file = audio_files[choice]
audio = AudioSegment.from_file(audio_file)

# Play audio using IPython.display

return ipd.Audio(data=audio_file, rate=audio.frame_rate)

# Example usage:
if __name__ == "__main__":
audio_files = [
"/music.wav",
"/oboe.wav",
"/fantastic.wav"
]

audio_widget = play_audio(audio_files)
if audio_widget:
display(audio_widget)

https://colab.research.google.com/drive/10hNM-QiWZCSknm9v49nXRXHuJaHWEML4?authuser=3#scrollTo=la4RB8soOwhM&printMode=true 5/7
7/22/24, 2:51 PM DSPA_ET22BTEC046_LAB3.ipynb - Colab

Choose an audio file to play:

1: /music.wav
2: /oboe.wav
3: /fantastic.wav
Enter your choice (1-3): 2

0:01 / 0:02

(e) Input the Speech signal using microphone and perform the above process

import IPython.display as ipd

from pydub import AudioSegment

audio_file = "/pranil.wav"

# Load audio file

audio = AudioSegment.from_file(audio_file)

# Play audio using IPython.display

ipd.Audio(data=audio_file, rate=audio.frame_rate)

0:02 / 0:04

from pydub import AudioSegment

import numpy as np
import matplotlib.pyplot as plt
audio_file = "/pranil.wav"
audio = AudioSegment.from_file(audio_file)
duration_ms = len(audio)
print(f"Duration of the audio: {duration_ms} milliseconds")

samples = audio.get_array_of_samples()

# Calculate duration (in seconds) and sample rate

duration = len(samples) / audio.frame_rate
sample_rate = audio.frame_rate

# Create a time vector spaced linearly with the duration of the audio
time = np.linspace(0, duration, num=len(samples))

# Plotting the waveform

plt.figure(figsize=(10, 4))
plt.plot(time, samples, color='b')
plt.xlabel('Time (s)')
plt.ylabel('Amplitude')
plt.title('Audio Waveform')
plt.grid(True)
plt.show()

https://colab.research.google.com/drive/10hNM-QiWZCSknm9v49nXRXHuJaHWEML4?authuser=3#scrollTo=la4RB8soOwhM&printMode=true 6/7
7/22/24, 2:51 PM DSPA_ET22BTEC046_LAB3.ipynb - Colab

Duration of the audio: 4800 milliseconds

from pydub import AudioSegment

import numpy as np
import matplotlib.pyplot as plt
from scipy import signal

# List of audio files

audio_files = [
"/pranil.wav"
]

# Initialize lists to store durations and samples of each audio file

durations = []
samples_list = []

# Loop through each audio file

for audio_file in audio_files:
# Load audio file
audio = AudioSegment.from_file(audio_file)

# Calculate duration (in seconds) and sample rate

duration = len(audio) / 1000 # Convert duration to seconds
sample_rate = audio.frame_rate

# Extract raw audio data (as numpy array)

samples = np.array(audio.get_array_of_samples())

# Store duration and samples

durations.append(duration)
samples_list.append(samples)

https://colab.research.google.com/drive/10hNM-QiWZCSknm9v49nXRXHuJaHWEML4?authuser=3#scrollTo=la4RB8soOwhM&printMode=true 7/7