gitphp 0.2.9.1 :: scannr.git/commitdiff

add some pyaudio test files

3 files changed: (show all)
snd.py
test.py (new)
test2.py (new)

file:a/snd.py -> file:b/snd.py

""" Record a few seconds of audio and save to a WAVE file.	""" Record a few seconds of audio and save to a WAVE file.
Based on http://stackoverflow.com/questions/892199/detect-record-audio-in-python/6743593#6743593	Based on http://stackoverflow.com/questions/892199/detect-record-audio-in-python/6743593#6743593
"""	"""

import pyaudio	import pyaudio
import wave	import wave
import sys	import sys
import audioop # http://docs.python.org/library/audioop	import audioop # http://docs.python.org/library/audioop
from os.path import exists	from os.path import exists
from array import array	from array import array
from struct import unpack, pack	from struct import unpack, pack
import threading	import threading
from pydispatch import dispatcher	from pydispatch import dispatcher

THRESHOLD = 500	THRESHOLD = 500
CHUNK_SIZE = 1024	CHUNK_SIZE = 1024
FORMAT = pyaudio.paInt16	FORMAT = pyaudio.paInt16
RATE = 44100	RATE = 44100
if sys.platform.startswith('darwin'):	if sys.platform.startswith('darwin'):
CHANNELS = 2	CHANNELS = 2
elif sys.platform.startswith('win32'):	elif sys.platform.startswith('win32'):
CHANNELS = 1	CHANNELS = 1

MAX_SILENT = 30	MAX_SILENT = 30

def is_silent(L):	def is_silent(L):
"Returns `True` if below the 'silent' threshold"	"Returns `True` if below the 'silent' threshold"
"print max(L)"	"print max(L)"
"print max(L) < THRESHOLD"	"print max(L) < THRESHOLD"
return max(L) < THRESHOLD	return max(L) < THRESHOLD

def normalize(L):	def normalize(L):
"Average the volume out"	"Average the volume out"
MAXIMUM = 16384	MAXIMUM = 16384
times = float(MAXIMUM)/max(abs(i) for i in L)	times = float(MAXIMUM)/max(abs(i) for i in L)

LRtn = array('h')	LRtn = array('h')
for i in L:	for i in L:
LRtn.append(int(i*times))	LRtn.append(int(i*times))
return LRtn	return LRtn

def trim(L):	def trim(L):
"Trim the blank spots at the start and end"	"Trim the blank spots at the start and end"
def _trim(L):	def _trim(L):
snd_started = False	snd_started = False
LRtn = array('h')	LRtn = array('h')

for i in L:	for i in L:
if not snd_started and abs(i)>THRESHOLD:	if not snd_started and abs(i)>THRESHOLD:
snd_started = True	snd_started = True
LRtn.append(i)	LRtn.append(i)

elif snd_started:	elif snd_started:
LRtn.append(i)	LRtn.append(i)
return LRtn	return LRtn

# Trim to the left	# Trim to the left
L = _trim(L)	L = _trim(L)

# Trim to the right	# Trim to the right
L.reverse()	L.reverse()
L = _trim(L)	L = _trim(L)
L.reverse()	L.reverse()
return L	return L

def add_silence(L, seconds):	def add_silence(L, seconds):
"Add silence to the start and end of `L` of length `seconds` (float)"	"Add silence to the start and end of `L` of length `seconds` (float)"
LRtn = array('h', [0 for i in xrange(int(seconds*RATE))])	LRtn = array('h', [0 for i in xrange(int(seconds*RATE))])
LRtn.extend(L)	LRtn.extend(L)
LRtn.extend([0 for i in xrange(int(seconds*RATE))])	LRtn.extend([0 for i in xrange(int(seconds*RATE))])
return LRtn	return LRtn

def record():	def record():
"""	"""
Record a word or words from the microphone and	Record a word or words from the microphone and
return the data as an array of signed shorts.	return the data as an array of signed shorts.

Normalizes the audio, trims silence from the	Normalizes the audio, trims silence from the
start and end, and pads with 0.5 seconds of	start and end, and pads with 0.5 seconds of
blank sound to make sure VLC et al can play	blank sound to make sure VLC et al can play
it without getting chopped off.	it without getting chopped off.
"""	"""
p = pyaudio.PyAudio()	p = pyaudio.PyAudio()
stream = p.open(format=FORMAT, channels=CHANNELS, rate=RATE,	stream = p.open(format=FORMAT, channels=CHANNELS, rate=RATE,
input=True,	input=True,
frames_per_buffer=CHUNK_SIZE)	frames_per_buffer=CHUNK_SIZE)

num_silent = 0	num_silent = 0
snd_started = False	snd_started = False

LRtn = array('h')	LRtn = array('h')

while 1:	while 1:
data = stream.read(CHUNK_SIZE)	data = stream.read(CHUNK_SIZE)
L = unpack('<' + ('h'*(len(data)/2)), data) # little endian, signed short	L = unpack('<' + ('h'*(len(data)/2)), data) # little endian, signed short
L = array('h', L)	L = array('h', L)

silent = is_silent(L)	silent = is_silent(L)
#print silent, num_silent, L[:10]	#print silent, num_silent, L[:10]

if silent and snd_started:	if silent and snd_started:
num_silent += 1	num_silent += 1
print num_silent	print num_silent
elif not silent and not snd_started:	elif not silent and not snd_started:
dispatcher.send( signal='SND_STARTED')	dispatcher.send( signal='SND_STARTED')
snd_started = True	snd_started = True
print snd_started	print snd_started
if snd_started:	if snd_started:
LRtn.extend(L)	LRtn.extend(L)
if snd_started and num_silent > MAX_SILENT:	if snd_started and num_silent > MAX_SILENT:
break	break

sample_width = p.get_sample_size(FORMAT)	sample_width = p.get_sample_size(FORMAT)
stream.stop_stream()	stream.stop_stream()
stream.close()	stream.close()
p.terminate()	p.terminate()

LRtn = normalize(LRtn)	LRtn = normalize(LRtn)
LRtn = trim(LRtn)	LRtn = trim(LRtn)
LRtn = add_silence(LRtn, 0.5)	LRtn = add_silence(LRtn, 0.5)
return sample_width, LRtn	return sample_width, LRtn

def record_to_file(path):	def record_to_file(path):
"Records from the microphone and outputs the resulting data to `path`"	"Records from the microphone and outputs the resulting data to `path`"
sample_width, data = record()	sample_width, data = record()
data = pack('<' + ('h'len(data)), data)	data = pack('<' + ('h'len(data)), data)

wf = wave.open(path, 'wb')	wf = wave.open(path, 'wb')
wf.setnchannels(1)	wf.setnchannels(CHANNELS)
wf.setsampwidth(sample_width)	wf.setsampwidth(sample_width)
wf.setframerate(RATE)	wf.setframerate(RATE)
wf.writeframes(data)	wf.writeframes(data)
wf.close()	wf.close()
print("done - result written to "+path)	print("done - result written to "+path)
del data	del data




if __name__ == '__main__':	if __name__ == '__main__':
filename = 'demo.wav'	filename = 'demo.wav'
record_to_file(filename)	record_to_file(filename)
print("done - result written to "+filename)	print("done - result written to "+filename)

file:b/test.py (new)

	"""PyAudio example: Record a few seconds of audio and save to a WAVE file."""

	import pyaudio
	import wave

	CHUNK = 1024
	FORMAT = pyaudio.paInt16
	CHANNELS = 2
	RATE = 44100
	RECORD_SECONDS = 5
	WAVE_OUTPUT_FILENAME = "output.wav"

	p = pyaudio.PyAudio()

	stream = p.open(format=FORMAT,
	channels=CHANNELS,
	rate=RATE,
	input=True,
	frames_per_buffer=CHUNK)

	print("* recording")

	frames = []

	for i in range(0, int(RATE / CHUNK * RECORD_SECONDS)):
	data = stream.read(CHUNK)
	frames.append(data)

	print("* done recording")

	stream.stop_stream()
	stream.close()
	p.terminate()

	wf = wave.open(WAVE_OUTPUT_FILENAME, 'wb')
	wf.setnchannels(CHANNELS)
	wf.setsampwidth(p.get_sample_size(FORMAT))
	wf.setframerate(RATE)
	wf.writeframes(b''.join(frames))
	wf.close()

file:b/test2.py (new)

	from os.path import exists
	from array import array
	from struct import unpack, pack

	import pyaudio
	import wave

	THRESHOLD = 500
	CHUNK_SIZE = 1024
	FORMAT = pyaudio.paInt16
	RATE = 44100

	def is_silent(L):
	"Returns `True` if below the 'silent' threshold"
	return max(L) < THRESHOLD

	def normalize(L):
	"Average the volume out"
	MAXIMUM = 16384
	times = float(MAXIMUM)/max(abs(i) for i in L)

	LRtn = array('h')
	for i in L:
	LRtn.append(int(i*times))
	return LRtn

	def trim(L):
	"Trim the blank spots at the start and end"
	def _trim(L):
	snd_started = False
	LRtn = array('h')

	for i in L:
	if not snd_started and abs(i)>THRESHOLD:
	snd_started = True
	LRtn.append(i)

	elif snd_started:
	LRtn.append(i)
	return LRtn

	# Trim to the left
	L = _trim(L)

	# Trim to the right
	L.reverse()
	L = _trim(L)
	L.reverse()
	return L

	def add_silence(L, seconds):
	"Add silence to the start and end of `L` of length `seconds` (float)"
	LRtn = array('h', [0 for i in xrange(int(seconds*RATE))])
	LRtn.extend(L)
	LRtn.extend([0 for i in xrange(int(seconds*RATE))])
	return LRtn

	def record():
	"""
	Record a word or words from the microphone and
	return the data as an array of signed shorts.

	Normalizes the audio, trims silence from the
	start and end, and pads with 0.5 seconds of
	blank sound to make sure VLC et al can play
	it without getting chopped off.
	"""
	p = pyaudio.PyAudio()
	stream = p.open(format=FORMAT, channels=2, rate=RATE,
	input=True,
	frames_per_buffer=CHUNK_SIZE)

	num_silent = 0
	snd_started = False

	LRtn = array('h')

	while 1:
	data = stream.read(CHUNK_SIZE)
	L = unpack('<' + ('h'*(len(data)/2)), data) # little endian, signed short
	L = array('h', L)
	LRtn.extend(L)

	silent = is_silent(L)
	#print silent, num_silent, L[:10]

	if silent and snd_started:
	num_silent += 1
	elif not silent and not snd_started:
	snd_started = True

	if snd_started and num_silent > 30:
	break

	sample_width = p.get_sample_size(FORMAT)
	stream.stop_stream()
	stream.close()
	p.terminate()

	LRtn = normalize(LRtn)
	LRtn = trim(LRtn)
	LRtn = add_silence(LRtn, 0.5)
	return sample_width, LRtn

	def record_to_file(path):
	"Records from the microphone and outputs the resulting data to `path`"
	sample_width, data = record()
	data = pack('<' + ('h'len(data)), data)

	wf = wave.open(path, 'wb')
	wf.setnchannels(2)
	wf.setsampwidth(sample_width)
	wf.setframerate(RATE)
	wf.writeframes(data)
	wf.close()

	if __name__ == '__main__':
	print("please speak a word into the microphone")
	record_to_file('demo.wav')
	print("done - result written to demo.wav")