Generating wave sound from text file

Question

I have generated sequence of frequency sound from text file using :

import mmap
import math
import pyaudio

fh = open('/home/jay/Documents/try.txt', 'rb')

m = mmap.mmap(fh.fileno(), 0, access=mmap.ACCESS_READ)
ba = bytearray(m)

#sudo apt-get install python-pyaudio
PyAudio = pyaudio.PyAudio

#See http://en.wikipedia.org/wiki/Bit_rate#Audio
BITRATE = 16000 #number of frames per second/frameset.      

for freq in ba:
#See http://www.phy.mtu.edu/~suits/notefreqs.html
   FREQUENCY = 300 + freq #Hz, waves per second, 261.63=C4-note.
   LENGTH = 1 #seconds to play sound

   NUMBEROFFRAMES = int(BITRATE * LENGTH)
   RESTFRAMES = NUMBEROFFRAMES % BITRATE
   WAVEDATA = ''    

   for x in xrange(NUMBEROFFRAMES):
    WAVEDATA = WAVEDATA+chr(int(math.sin(x/((BITRATE/FREQUENCY)/math.pi))*127+128))    

   #fill remainder of frameset with silence
   for x in xrange(RESTFRAMES): 
    WAVEDATA = WAVEDATA+chr(128)

   p = PyAudio()
   stream = p.open(format = p.get_format_from_width(1), 
                   channels = 1, 
                   rate = BITRATE, 
                   output = True)
   stream.write(WAVEDATA)
   stream.stop_stream()
   stream.close()
   p.terminate()

(try.txt can be any text file you want)

But its having some noise in between frequency sound how can i remove it and save sequence of all frequncy played in .wave or .mp3 file? Sorry i am still learning so if i am not clear in asking. Thanks, Jay

I don't hear anything suspicious (I used python 3 and adapted your code). Can you share you txt file? (would have been better with a hardcoded list for this almost [mcve]) — Jean-François Fabre, Sep 10 '16 at 12:34
hi i used python2.7 also pyaudio. for text file i just copied same code to it. Thanks. — Jay Patel, Sep 10 '16 at 14:12
This is my try.txt (http://pastebin.com/tnWKTdjZ) same as code i am using to produce sound. Thanks. — Jay Patel, Sep 11 '16 at 09:43
Jean that is the content of try.txt ... using same code as text into .txt file for generating sound. or you can just copy paste any text into try.txt Thanks. — Jay Patel, Sep 11 '16 at 09:57
hi Jean it is like this : http://pasteall.org/pic/index.php?id=106597 sorry if my english is rusty. Thanks for being patient with me — Jay Patel, Sep 11 '16 at 10:07
yes jean.. code working properly if you try it. but only thing is there's two dot tones in between frequency sounds... i can't figure out that is causing them and want to record output to try.wave file — Jay Patel, Sep 11 '16 at 12:11
If I understand correctly, you send a lot of quasi-random frequencies to your program and expect it to sound correct/predictable? Ok I get it. It's fun. And I get a clicking sound when inserting a silence. That's your problem right? — Jean-François Fabre, Sep 11 '16 at 16:02

Jean-François Fabre · Accepted Answer · 2016-09-11T16:17:34.220

Biggest problem here is that you open/close audio stream at each frequency.

Instead, keep the stream open and you'll get a lot less "clicks". That doesn't mean you'll get no clicks, probably because processing time is sometimes too long and the stream is interrupted. The best way would be to multithread generation and write to be more reactive, but that's another story...

I also added .wav save capacity. My output filenames. Works very well.

Fixed code (runs with Python 3)

import mmap
import math
import pyaudio,wave
import array


fh = open('K:\out.txt', 'rb')

m = mmap.mmap(fh.fileno(), 0, access=mmap.ACCESS_READ)
ba = bytearray(m)

#ba = [300,400,500,400,200]
#sudo apt-get install python-pyaudio
PyAudio = pyaudio.PyAudio

#See http://en.wikipedia.org/wiki/Bit_rate#Audio
BITRATE = 16000 #number of frames per second/frameset.

p = PyAudio()
stream = p.open(format = p.get_format_from_width(1),
               channels = 1,
               rate = BITRATE,
               output = True)

wf=wave.open("K:\wavout.wav","wb")
wf.setnchannels(1)
wf.setsampwidth(p.get_sample_size(pyaudio.paInt8))  # byte = 8 bits else trashed
wf.setframerate(BITRATE)


for freq in ba:
#See http://www.phy.mtu.edu/~suits/notefreqs.html
   FREQUENCY = 300 + freq #Hz, waves per second, 261.63=C4-note.
   print('freq '+str(FREQUENCY))
   LENGTH = 1 #seconds to play sound

   NUMBEROFFRAMES = int(BITRATE * LENGTH)
   RESTFRAMES = NUMBEROFFRAMES % BITRATE
   WAVEDATA = list()


   for x in range(NUMBEROFFRAMES):
    v = int(math.sin(x/((BITRATE/FREQUENCY)/math.pi))*127+128)
    WAVEDATA.append(v)

   #fill remainder of frameset with silence
   WAVEDATA+=[128]*RESTFRAMES

   b = array.array('B', WAVEDATA).tostring()

   wf.writeframes(b)
   stream.write(b)

#print data
wf.close()
stream.stop_stream()
stream.close()
p.terminate()

thanks its working .. have some clicks because of the issue you mentioned.. but it works like charm for me :) Thanks a ton .. — Jay Patel, Sep 12 '16 at 17:03

Generating wave sound from text file

1 Answers1