Pyaudio mix mp3 files with microphone

Question

I am trying to mix 2 mp3 files with microphone input using pydub and pyaudio. Here is what i have done:

from pydub import AudioSegment
from pydub.playback import play

sound1 = AudioSegment.from_file(r"ΑΓΙΑ ΣΚΕΠΗ.mp3")
sound1_channels = sound1.split_to_mono()
sound1 = sound1_channels[0].overlay(sound1_channels[1])
sound1 = sound1 - 48 # make sound1 quiter 48dB

sound2 = AudioSegment.from_file(r"ΑΓΙΑ ΚΥΡΙΑΚΗ.mp3")
sound2_channels = sound2.split_to_mono()
sound2 = sound2_channels[0].overlay(sound2_channels[1])
sound2 = sound2 - 48 # make sound2 quiter 48dB

import pyaudio

FORMAT = pyaudio.paInt16
CHANNELS = 1
RATE = 44100
CHUNK = 1024

p = pyaudio.PyAudio()
player = p.open(format=FORMAT, channels=CHANNELS, rate=RATE, output=True,frames_per_buffer=CHUNK)
mic_stream = p.open(format=FORMAT, channels=CHANNELS,rate=RATE, input=True,frames_per_buffer=CHUNK)

chunk_time_in_seconds = int(RATE/CHUNK)
chunk_number = 0

while(True):
    mic_data = mic_stream.read(CHUNK)
    mic_sound = AudioSegment(mic_data, sample_width=2, channels=1, frame_rate=RATE)
    
    sound1_part = sound1[chunk_number*chunk_time_in_seconds:(chunk_number+1)*chunk_time_in_seconds]
    sound2_part = sound2[chunk_number*chunk_time_in_seconds:(chunk_number+1)*chunk_time_in_seconds]
    
    
    
    #player.write(mic_sound.raw_data) works well
    mix_sound = sound1_part.overlay(sound2_part).overlay(mic_sound)
    player.write(mix_sound.raw_data) # low microphone quality
    
    chunk_number = chunk_number+1

The above code seems to work, but it output microphone input with low quality.

Can you please help me about?

Does it work if you mix three mp3? This could help to determine if the issue is specific to the mic input or is it possibly the mixing method... — voyager42, Feb 24 '21 at 16:17

Chris P · Accepted Answer · 2021-02-24T22:55:58.473

I don't know if this aproach is good for real time audio processing but the following code works well!

from pydub import AudioSegment
from pydub.playback import play

sound1 = AudioSegment.from_file(r"ΑΓΙΑ ΣΚΕΠΗ.mp3")
sound1_channels = sound1.split_to_mono()
sound1 = sound1_channels[0].overlay(sound1_channels[1])
sound1 = sound1 - 30 # make sound1 quiter 30dB

sound2 = AudioSegment.from_file(r"ΑΓΙΑ ΚΥΡΙΑΚΗ.mp3")
sound2_channels = sound2.split_to_mono()
sound2 = sound2_channels[0].overlay(sound2_channels[1])
sound2 = sound2 - 30 # make sound2 quiter 30dB

import pyaudio

FORMAT = pyaudio.paInt16
CHANNELS = 1
RATE = 44100
CHUNK = 1024

p = pyaudio.PyAudio()
player = p.open(format=FORMAT, channels=CHANNELS, rate=RATE, output=True,frames_per_buffer=CHUNK)
mic_stream = p.open(format=FORMAT, channels=CHANNELS,rate=RATE, input=True,frames_per_buffer=CHUNK)

#chunk_time_in_seconds = int(RATE/CHUNK)
chunk_number = 0

while(True):
    mic_data = mic_stream.read(CHUNK)
    mic_sound = AudioSegment(mic_data, sample_width=2, channels=1, frame_rate=RATE)
    mic_sound_duration = len(mic_sound)
    
    sound1_part = sound1[chunk_number*mic_sound_duration:(chunk_number+1)*mic_sound_duration]
    sound2_part = sound2[chunk_number*mic_sound_duration:(chunk_number+1)*mic_sound_duration]
    
    
    
    #player.write(mic_sound.raw_data) works well
    mix_sound = sound1_part.overlay(sound2_part).overlay(mic_sound)
    player.write(mix_sound.raw_data) # low microphone quality
    
    chunk_number = chunk_number+1

Pyaudio mix mp3 files with microphone

1 Answers1