def savewav(self, filename):
wf = wave.open(filename, 'wb')
wf.setnchannels(1)
wf.setsampwidth(2)
wf.setframerate(self.SAMPLING_RATE)
wf.writeframes(np.array(self.Voice_String).tostring())
wf.close()
def write_wav(data, filename, framerate, amplitude, length):
wavfile = wave.open(filename,'w')
nchannels = 1
sampwidth = 2
framerate = framerate
nframes = length
comptype = "NONE"
compname = "not compressed"
wavfile.setparams((nchannels,
sampwidth,
framerate,
nframes,
comptype,
compname))
frames = []
for s in data:
mul = int(s * amplitude)
print("I am here")
print(mul)
frames.append(struct.pack('h', mul))
frames = ''.join(frames)
wavfile.writeframes(frames)
wavfile.close()
print("%s written" %(filename))
def save_file(filename):
wavfile = wave.open(filename, "w")
# wav parameters
# Number of channels
nchannels = 1
# Sample width in bytes
sampwidth = 2
# Number of frames or samples
nframes = len(data)
comptype = "NONE"
compname = "not compressed"
wavfile.setparams(
(nchannels, sampwidth, samples, nframes, comptype, compname))
for sample in data:
"""
struct - Takes the data from my "data" list and packs it as binary data or else
the file is not readable by an audio player.
h - Means 16 bit number
32767.0 - The floating point numbers are not represented right and will not work
when writing to the wave file. Convert the floating point numbers to fixed point
to get full scale audio.
"""
wavfile.writeframes(struct.pack('h', int(sample * 32767.0)))
wavfile.close()
return
def load(self, originalname):
# Load + convert audio
print("Loading sound file <{}>...".format(originalname))
name = originalname.split(".")[0]
ext = originalname.split(".")[-1].lower()
audiofile = originalname
if ext == MP3:
newfilename = name + ".wav"
old = AudioSegment.from_mp3(audiofile)
old.export(newfilename, format=WAV)
self.filename = newfilename
elif ext == PCM:
print "This is a PCM file!"
pcmfile = open(audiofile, 'rb')
pcmdata = pcmfile.read()
pcmfile.close()
wavfile = wave.open(name + '.wav', 'wb')
wavfile.setparams((1, 2, 44100, 16, 'NONE', 'NONE'))
wavfile.writeframes(pcmdata)
wavfile.close()
self.filename = name + ".wav"
else:
self.filename = audiofile
def get_wav_info(wav_file):
wav = wave.open(wav_file, 'r')
frames = wav.readframes(-1)
sound_info = pylab.frombuffer(frames, 'int16')
frame_rate = wav.getframerate()
wav.close()
return sound_info, frame_rate
def duration(wav_file):
wav = wave.open(wav_file, 'r')
frames = wav.getnframes()
rate = wav.getframerate()
dur = frames / float(rate)
#print 'duration: %f' % duration
wav.close()
return dur
def write_text_grid_from_hash(grid_name, seg, sample_rate, len_signal, levels):
try:
w = None
w = open(grid_name, "w", encoding="utf8")
if w is None:
return
#chapka
w.write(
"File type = \"ooTextFile\"\nObject class = \"TextGrid\"\n\nxmin = 0.0\n"
)
xmax = len_signal / sample_rate
w.write("xmax = {0}\ntiers? <exists>\nsize = {1}\nitem []:\n".format(
float(xmax), len(seg)))
size_total = 0
ind_level = 1
for level_name in levels:
try:
cur_intervals = seg[level_name]
w.write("\titem [{0}]\n".format(ind_level))
w.write("\t\tclass = \"IntervalTier\"\n")
w.write("\t\tname = \"{0}\"\n".format(level_name))
w.write("\t\txmin = {0}\n".format(
float(cur_intervals[0]["frm"] / sample_rate)))
w.write("\t\txmax = {0}\n".format(float(xmax)))
if len(cur_intervals):
w.write("\t\tintervals: size = {0}\n".format(
len(cur_intervals)))
j_ind = 1
for item in cur_intervals:
w.write("\t\tintervals [{0}]:\n".format(j_ind))
w.write("\t\t\txmin = {0}\n".format(
float(item["frm"] / sample_rate)))
w.write("\t\t\txmax = {0}\n".format(
float(item["to"] / sample_rate)))
w.write("\t\t\ttext = \"{0}\"\n".format(item["nm"]))
j_ind += 1
else:
w.write("\t\tintervals: size = 1\n")
w.write("\t\tintervals [1]:\n")
w.write("\t\t\txmin = 0\n")
w.write("\t\t\txmax = {0}\n".format(
float(item["to"] / sample_rate)))
w.write("\t\t\ttext = \"\"\n")
a = 1
except Exception as err:
print(err)
except Exception as err:
print(err)
finally:
if w is not None:
w.close()
def capture_audio(callback, final_callback):
import pyaudio
import wave as wv
from array import array
FORMAT = pyaudio.paInt16
RATE = 16000
CHUNK = 1024
COUNTCHUNK = (RATE + CHUNK - 1) / CHUNK
RECORD_SECONDS = 1
FILE_NAME = "/tmp/mic_rec.wav"
audio = pyaudio.PyAudio()
stream = audio.open(format=FORMAT,
rate=RATE,
channels=1,
input=True,
frames_per_buffer=CHUNK)
#starting recording
wave = []
callback_free = [True]
print("started")
last = time.time()
lastloudsoundtime = time.time() - 100000
while True:
data = stream.read(CHUNK)
data_chunk = array('h', data)
vol = max(data_chunk)
# print("Volume {}".format(vol))
wave.append(data)
if len(wave) > COUNTCHUNK:
wave.pop(0)
if (vol >= 2500):
lastloudsoundtime = time.time()
if (time.time() - last > 0.05
and time.time() - lastloudsoundtime < 0.8):
last = time.time()
if len(wave) == COUNTCHUNK:
#writing to file
wavfile = wv.open(FILE_NAME, 'wb')
wavfile.setnchannels(1)
wavfile.setsampwidth(audio.get_sample_size(FORMAT))
wavfile.setframerate(RATE)
wavfile.writeframes(b''.join(wave))
wavfile.close()
if callback_free[0]:
callback_free[0] = False
callback(callback_free, FILE_NAME, final_callback)
# print("Something is said")
else:
pass
def read_wav_data(filename):
wav = wave.open(filename, "rb") # 打开一个wav格式的声音文件流
num_frame = wav.getnframes() # 获取帧数
num_channel = wav.getnchannels() # 获取声道数
framerate = wav.getframerate() # 获取帧速率
num_sample_width = wav.getsampwidth() # 获取实例的比特宽度，即每一帧的字节数
str_data = wav.readframes(num_frame) # 读取全部的帧
wav.close() # 关闭流
wave_data = np.fromstring(str_data, dtype=np.short) # 将声音文件数据转换为数组矩阵形式
wave_data.shape = -1, num_channel # 按照声道数将数组整形，单声道时候是一列数组，双声道时候是两列的矩阵
wave_data = wave_data.T # 将矩阵转置
return wave_data, framerate
def save_wave_file(filename, data):
# define of params
framerate = 8000
channels = 1
sampwidth = 2
'''save the date to the wav file'''
wf = wave.open(filename, 'wb')
wf.setnchannels(channels)
wf.setsampwidth(sampwidth)
wf.setframerate(framerate)
wf.writeframes("".join(data))
wf.close()
return 0
def stop(self, path='output.wav'):
sd.stop()
duration = int(math.ceil(time.time() - self._start_time))
data = self._recording[:duration * sd.default.samplerate]
# save audio
audio = pyaudio.PyAudio()
format = pyaudio.paInt16
wavfile = wave.open(path, 'wb')
wavfile.setnchannels(CHANNELS)
wavfile.setsampwidth(audio.get_sample_size(format))
wavfile.setframerate(SAMPLERATE)
wavfile.writeframes(data.tostring())
wavfile.close()
def write_wav(data, filename, framerate, amplitude):
wavfile = wave.open(filename,'w')
nchannels = 1
sampwidth = 2
framerate = framerate
nframes = len(data)
comptype = "NONE"
compname = "not compressed"
wavfile.setparams((nchannels,
sampwidth,
framerate,
nframes,
comptype,
compname))
frames = []
for s in data:
mul = int(s * amplitude)
frames.append(struct.pack('h', mul))
frames = ''.join(frames)
wavfile.writeframes(frames)
wavfile.close()
print("%s written" %(filename))
frames = []
for i in range(0, int(RATE / CHUNK * RECORD_SECONDS)):
data = stream.read(CHUNK)
frames.append(data)
stream.stop_stream()
stream.close()
audio.terminate()
wavfile = wave.open(FILE_NAME, 'w')
wavfile.setnchannels(CHANNELS)
wavfile.setsampwidth(audio.get_sample_size(FORMAT))
wavfile.setframerate(RATE)
wavfile.writeframes(b''.join(frames))
wavfile.close()
wavfile = wave.open(FILE_NAME, 'r')
samplingFrequency, signalData = scipy.io.wavfile.read('ses.wav')
plot.subplot(211)
plot.plot(signalData)
plot.subplot(212)
plot.specgram(signalData, Fs=samplingFrequency)
plot.show()
#3.Kısım - Real time mikrofon ses verileriyle sinyal işleme
import pyaudio
import struct
import numpy as np
import matplotlib.pyplot as plt
hexString += (hex(binVal)[2:-1])
decryptMess = binascii.unhexlify(hexString)
print("original text: " +decryptMess)
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()
#-------------------------- ----------------------------
#plotting sound below
#------------------------------------------------------
spf = wave.open('input.wav','r')
signal = spf.readframes(-1)
signal = np.fromstring(signal, 'Int16')
fs = spf.getframerate()
#If Stereo
if spf.getnchannels() == 2:
print 'Just mono files'
sys.exit(0)