Create a Voice Recorder using Python

---

---

Here, we will create an application (voice recorder) using Python that will allow the user to record voice. We will create this voice recorder application using the sounddevice library and Python.

By using the voice recorder, you can initiate the recording process, pause it or, if necessary, cancel. The recorded audio is saved in two formats WAV and WAVIO. This voice recorder application can be useful for such application that needs only plain audio record and playback.

Required Libraries

To create voice recorder in Python, the following are the required libraries −

• sounddevice − The sounddevice library will be used for capturing audio.
• scipy − This library will be used for writing WAV files using write.
• wavio − This library will be used for writing wav files with other options.
• numpy − This library will be used for matrices or array calculations.

Installation

To install the required libraries to create voice records, you need to install these libraries using the PIP. Use the following statement to install all of the required libraries at once −

pip install sounddevice scipy wavio numpy

Steps to Create a Voice Recorder

1. Import Libraries

After the installation, you need to import all libraries in the code. To import the required libraries, use the following code statements −

import sounddevice as sd
from scipy.io.wavfile import write
import wavio as wv
import numpy as np
import threading

Step 2: Initialize Variables

Use the following code to set up the sampling frequency and flags for managing the recording state −

• To set sampling frequency −

freq = 44100

• To store the recorded audio segments −

recording = []

• Set the default flag value to check whether recording is active or not −

is_recording = False

• Set the default flag value to stop recording −

stop_recording = False

Step 3: Define Recording Function

Create a function to handle audio recording in a loop, appending each segment to the recording list. Use the following code −

def record_audio():
   global recording, is_recording, stop_recording
   is_recording = True
   while is_recording and not stop_recording:
      recording_segment = sd.rec(int(freq), samplerate=freq, channels=2)
      sd.wait()  # Wait until the recording is finished
      recording.append(recording_segment)

Step 4: Define Input Handling Function

Implement a function to process user commands for recording control (start, pause, stop). Use the following code −

def handle_input():
   global is_recording, stop_recording
   while True:
      user_input = input("r to START , p to PAUSE and s to STOP the recording, please choice accordingly \n Enter command: ").lower()
      if user_input == 'p':
         if is_recording:
            print("Recording Paused, (you can again start the recording using 'r')")
            is_recording = False
         else:
            print("Recording is not active.")
      elif user_input == 'r':
         if not is_recording:
            print("Recording Resumed (type 'p' to pause or 's' to stop)")
            threading.Thread(target=record_audio).start()
         else:
            print("Already recording.")
      elif user_input == 's':
         if is_recording:
            print("Stopping recording. Thank you for recording :)")
            is_recording = False
            stop_recording = True
            break
         else:
            print("Nothing to stop.")

Step 5: Combine and Save Recordings

After stopping the recording, combine all recorded segments into a single audio file and save it in different formats. Use the following code −

if recording:
   final_recording = np.concatenate(recording, axis=0)
   write("recording0.wav", freq, final_recording)
   wv.write("recording1.wav", final_recording, freq, sampwidth=2)
   print("Recording saved as 'recording0.wav' and 'recording1.wav'.")
else:
   print("No recording was made.")

Python Code to Create a Voice Recorder

Here is the complete code to create a voice recorder application in Python −

import sounddevice as sd
from scipy.io.wavfile import write
import wavio as wv
import numpy as np
import threading
# Sampling frequency
freq = 44100

# Initialize recording variables
recording = []
is_recording = False
stop_recording = False

def record_audio():
   global recording, is_recording, stop_recording
   is_recording = True
   while is_recording and not stop_recording:
      recording_segment = sd.rec(int(freq), samplerate=freq, channels=2)
      sd.wait()
      recording.append(recording_segment)

def handle_input():
   global is_recording, stop_recording
   while True:
      user_input = input("r to START , p to PAUSE and s to STOP the recording, please choice accordingly \n Enter command: ").lower()
      if user_input == 'p':
         if is_recording:
            print("Recording Paused, (you can again start the recording using 's') ")
            is_recording = False
         else:
            print("Recording is not working activly.")
      elif user_input == 'r':
         if not is_recording:
            print("Recording Resumed (type 'p' to pause or 's' to stop)")
            threading.Thread(target=record_audio).start()
         else:
            print("Already recording.")
      elif user_input == 's':
         if is_recording:
            print("Stopping recording. Thank you for recording :) ")
            is_recording = False
            stop_recording = True
            break
         else:
            print(" Nothing to stop.")

# Start the input handler in a separate thread
input_thread = threading.Thread(target=handle_input)
input_thread.start()

# Wait for the input thread to complete
input_thread.join()

# Combine all the recorded segments
if recording:
   final_recording = np.concatenate(recording, axis=0)
   write("recording0.wav", freq, final_recording)
   wv.write("recording1.wav", final_recording, freq, sampwidth=2)
   print("Recording saved as 'recording0.wav' and 'recording1.wav'.")
else:
   print("No recording was made.")

Output

After run this code you will see press "r" to start the recording, p to paused and s to stop the recording.

Now you can also see your recording audio file in your pc −

After that you can successfully listen your audio file −

Code Explanation

• Import Libraries − Import libraries for recording and processing audio as well as managing audio files.
• Initialize Variables − ‘Sampling rate’, ‘Recording list’ and ‘Flags’ as Recording control parameters should be set.
• Define record_audio Function − Records audio segments and adds it to the list.
• Define handle_input Function − Operates user’s instructions for beginning or stopping the recording process.
• Start Input Handling Thread − Manages the input handling function in a thread so that the game does not freeze.
• Wait for Input Thread to Complete − Waits for the input thread to complete the execution before proceeding with further operations of the main thread.
• Combine Recorded Segments − Combines all the recorded audio segments into one track or cue as they are called.
• Save Recording in WAV Format − Saves the combined audio as WAV files by two methods.
• Print Completion Message − Lets the users know that the recordings have been successfully saved.
• Handle No Recording Scenario − This one gives out an alarm if no audio was recorded.