def fuzz_effect():
RECORD_SECONDS = 10
BLOCKSIZE = 1024 # Number of frames per block
p = pyaudio.PyAudio()
WIDTH = 2 # bytes per sample
RATE = 44100 # Sampling rate (samples/second)
mix = 1
gain = 11
number_of_devices = p.get_device_count()
print ("There are {0:d} devices".format(number_of_devices))
property_list = ["defaultSampleRate", "maxInputChannels", "maxOutputChannels"]
for i in range(0, number_of_devices):
print ("Device {0:d} has:".format(i))
for s in property_list:
print " ", s, "=", p.get_device_info_by_index(i)[s]
stream = p.open(format=p.get_format_from_width(WIDTH), channels=2, rate=RATE, input=True, output=True)
output_block = [0.0 for n in range(0, 2 * BLOCKSIZE)]
q = 0.0
X = 0.0
y = 0.0
z = 0.0
r = 0.0
max_z = 0
max_r = 0
# k = 0.0
output_all = "" # output signal in all (string)
num_blocks = int(RATE / BLOCKSIZE * RECORD_SECONDS)
print ("* Playing...")
# Loop through wave file
for i in range(0, num_blocks):
# Get sample from wave file
input_string = stream.read(BLOCKSIZE)
# Convert string to number
input_value = struct.unpack("hh" * BLOCKSIZE, input_string)
X = np.fft.fft(input_value)
max_val = abs(np.max(X))
# print type(max_val)
max_val = max_val.item()
# print type(max_val)
# max_val = int(max_val)
for n in range(0, BLOCKSIZE):
x = input_value[2 * n]
q = x * gain / max_val
if q == 0:
z = 0
else:
z = -q / abs(q) * (1 - math.exp(-q * q / abs(q)))
if z > max_z:
max_z = z
r = mix * z * max_val / abs(max_z) + (1 - mix) * x
# k = np.asscalar(r)
if r > max_r:
max_r = r
if abs(max_r) == 0:
out = 0
else:
out = r * abs(x) / abs(max_r)
output_block[2 * n] = clip16_flt(out)
output_block[2 * n + 1] = clip16_flt(out)
output_string = struct.pack("hh" * BLOCKSIZE, *output_block)
# Write output to audio stream
stream.write(output_string)
output_all = output_all + output_string
print ("* Done")
stream.stop_stream()
stream.close()
p.terminate()
def fuzz_effect():
RECORD_SECONDS = 10
BLOCKSIZE = 1024 # Number of frames per block
p = pyaudio.PyAudio()
WIDTH = 2 # bytes per sample
RATE = 44100 # Sampling rate (samples/second)
mix = 1
gain = 11
number_of_devices = p.get_device_count()
print('There are {0:d} devices'.format(number_of_devices))
property_list = ['defaultSampleRate', 'maxInputChannels', 'maxOutputChannels']
for i in range(0, number_of_devices):
print('Device {0:d} has:'.format(i))
for s in property_list:
print ' ', s, '=', p.get_device_info_by_index(i)[s]
stream = p.open(format = p.get_format_from_width(WIDTH),
channels = 2,
rate = RATE,
input = True,
output = True)
output_block = [0.0 for n in range(0, 2*BLOCKSIZE)]
q = 0.0
X = 0.0
y = 0.0
z = 0.0
r = 0.0
max_z = 0
max_r = 0
# k = 0.0
output_all = '' # output signal in all (string)
num_blocks = int(RATE / BLOCKSIZE * RECORD_SECONDS)
print ('* Playing...')
# Loop through wave file
for i in range(0, num_blocks):
# Get sample from wave file
input_string = stream.read(BLOCKSIZE)
# Convert string to number
input_value = struct.unpack('hh'* BLOCKSIZE, input_string)
X = np.fft.fft(input_value)
max_val = abs(np.max(X))
# print type(max_val)
max_val = max_val.item()
# print type(max_val)
# max_val = int(max_val)
for n in range(0, BLOCKSIZE):
x = input_value[2*n]
q = x * gain / max_val
if q == 0:
z = 0
else:
z = -q/abs(q) * (1 - math.exp(-q*q/abs(q)))
if z>max_z:
max_z = z
r = mix * z * max_val/abs(max_z) + (1 - mix) * x
# k = np.asscalar(r)
if r>max_r:
max_r = r
if abs(max_r) ==0:
out = 0
else:
out = r * abs(x) / abs(max_r)
output_block[2*n] = clip16_flt(out)
output_block[2*n+1] = clip16_flt(out)
output_string = struct.pack('hh'* BLOCKSIZE , *output_block)
# Write output to audio stream
stream.write(output_string)
output_all = output_all + output_string
print('* Done')
stream.stop_stream()
stream.close()
p.terminate()
def fuzz_effect():
RECORD_SECONDS = 500
BLOCKSIZE = 1024 # Number of frames per block
p = pyaudio.PyAudio()
WIDTH = 2 # bytes per sample
RATE = 44100 # Sampling rate (samples/second)
stream = p.open(format = p.get_format_from_width(WIDTH),
channels = 2,
rate = RATE,
input = True,
output = True)
output_block = [0.0 for n in range(0, 2*BLOCKSIZE)]
q = 0.0
X = 0.0
y = 0.0
z = 0.000001
r = 0.0
max_z = 0.000001
max_r = 0
output_all = '' # output signal in all (string)
num_blocks = int(RATE / BLOCKSIZE * RECORD_SECONDS)
print ('##### playing Fuzzy #####')
# Loop through wave file
for i in range(0, num_blocks):
if app.MODE != "Fuzzy":
break
# Get sample from wave file
input_string = stream.read(BLOCKSIZE)
# Convert string to number
input_value = struct.unpack('hh'* BLOCKSIZE, input_string)
X = np.fft.fft(input_value)
max_val = abs(np.max(X))
max_val = max_val.item()
for n in range(0, BLOCKSIZE):
x = input_value[2*n]
q = x * parameters.fuzzy_gain / max_val
if q == 0:
z = 0
else:
z = -q/abs(q) * (1 - math.exp(-q*q/abs(q)))
if z > max_z:
max_z = z
r = parameters.fuzzy_mix * z * max_val / abs(max_z) + (1 - parameters.fuzzy_mix) * x
# k = np.asscalar(r)
if r > max_r:
max_r = r
if abs(max_r) ==0:
out = 0
else:
out = r * abs(x) / abs(max_r)
output_block[2*n] = clip16_flt(out)
output_block[2*n+1] = clip16_flt(out)
output_string = struct.pack('hh'* BLOCKSIZE , *output_block)
# Write output to audio stream
stream.write(output_string)
output_all = output_all + output_string
print('* Done')
stream.stop_stream()
stream.close()
p.terminate()
