这是学习时的笔记，包含相关资料链接，有的当时没有细看，记录下来在需要的时候回顾。
有些较混乱的部分，后续会再更新。
欢迎感兴趣的小伙伴一起讨论，跪求大神指点~
录音-语音流(pyAudio)
tags:voice
进行语音流检测
pyAudio
http://old.sebug.net/paper/books/scipydoc/wave_pyaudio.html
注：这部分是通过录音设备给语音激活检测传输语音流。
"path/to/vad/audio_stream.py"
#!usr/bin/env python
# -*- coding: utf-8 -*-
import numpy as np
from pyaudio import PyAudio,paInt16
from datetime import datetime
import wave
from Tkinter import *
import sys
from ffnn import FFNNVADGeneral
import logging
# import chardet # 查看编码
# define of params
NUM_SAMPLES =160
FRAMERATE = 16000
CHANNELS = 1
SAMPWIDTH = 2
FORMAT = paInt16
TIME = 125
FRAMESHIFT = 160
def save_wave_file(filename,data):
'''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)) # ""中间不能有空格，不然语音录入会有很多中断。
wf.close()
def my_button(root,label_text,button_text,button_stop,button_func,stop_func):
'''create label and button'''
label = Label(root,text=label_text,width=30,height=3).pack()
button = Button(root,text=button_text,command=button_func,anchor='center',width=30,height=3).pack()
button = Button(root,text=button_stop,command=stop_func,anchor='center',width=30,height=3).pack()
def record_wave():
'''open the input of wave'''
pa = PyAudio()
# 录音
stream = pa.open(format=FORMAT,
channels=CHANNELS,
rate=FRAMERATE,
input=True,
frames_per_buffer=NUM_SAMPLES) #一个buffer存NUM_SAMPLES个字节,作为一帧
vad = FFNNVADGeneral('/path/to/VAD/alex-master/alex/tools/vad_train/model_voip/vad_nnt_546_hu32_hl1_hla6_pf10_nf10_acf_1.0_mfr20000_mfl20000_mfps0_ts0_usec00_usedelta0_useacc0_mbo1_bs100.tffnn',
filter_length=2, sample_rate=16000, framesize=512, frameshift=160,
usehamming=True, preemcoef=0.97, numchans=26, ceplifter=22, numceps=12,
enormalise=True, zmeansource=True, usepower=True, usec0=False,
usecmn=False, usedelta=False, useacc=False, n_last_frames=10,
n_prev_frames=10, lofreq=125, hifreq=3800, mel_banks_only=True)
# 语音激活检测神经网络方法的类FFNNVADGeneral.
save_buffer = []
count = 0
# logging设置，用于记录日志
logging.basicConfig(level=logging.INFO,
filename='log.txt',
filemode ='w',
format='%(message)s')
while count < TIME*4:
string_audio_data = stream.read(NUM_SAMPLES)
result = vad.decide(string_audio_data)
frame = count*NUM_SAMPLES/float(FRAMESHIFT)
time = count*NUM_SAMPLES/float(FRAMERATE) # time=frame*frameshift/framerate
logging.info('frame: '+str(frame)+' time: '+str(time)+' prob: '+str(result)) # logging记录字符串，用‘+’连接
save_buffer.append(string_audio_data)
count += 1
#chardet.detect(string_audio_data) #查看编码类型
print "."
filename = datetime.now().strftime("%Y-%m-%d_%H_%M_%S")+".wav"
save_wave_file(filename,save_buffer)
save_buffer = []
print "filename,saved."
def record_stop():
# stop record the wave
sys.exit(0)
def main():
root = Tk()
root.geometry('300x200+200+200')
root.title('record wave')
my_button(root,"Record a wave","clik to record","stop recording",record_wave,record_stop)
root.mainloop()
if __name__ == "__main__":
main()
# error
$ bt_audio_service_open: connect() failed: Connection refused (111)
# 解决: 貌似有多余蓝牙库却没有蓝牙设备
$ sudo apt-get purge bluez-alsa
# Warning
$ ALSA lib pcm.c:2239:(snd_pcm_open_noupdate) Unknown PCM cards.pcm.rear
ALSA lib pcm.c:2239:(snd_pcm_open_noupdate) Unknown PCM cards.pcm.center_lfe
ALSA lib pcm.c:2239:(snd_pcm_open_noupdate) Unknown PCM cards.pcm.side
Cannot connect to server socket err = No such file or directory
Cannot connect to server request channel
jack server is not running or cannot be started
# 是由于usr/share/alsa/alsa.conf默认设置
录音
sudo apt-get update #更新软件源，最好使用国内的软件源，如何配置源参考百度。
sudo apt-get upgrade #升级软件包
sudo apt-get install alsa-utils alsa-tools alsa-tools-gui alsamixergui #安装所需软件包
# 查看音频设备
$ arecord -l
> card 0: PCH [HDA Intel PCH], device 0: ALC887-VD Analog [ALC887-VD Analog]
Subdevices: 1/1
Subdevice #0: subdevice #0
card 0: PCH [HDA Intel PCH], device 2: ALC887-VD Alt Analog [ALC887-VD Alt Analog]
Subdevices: 1/1
Subdevice #0: subdevice #0
# 机器有多于一个声卡，可以用下面的命令显示出来
$ cat /proc/asound/cards
> 0 [PCH ]: HDA-Intel - HDA Intel PCH
HDA Intel PCH at 0xf7210000 irq 29
1 [NVidia ]: HDA-Intel - HDA NVidia
HDA NVidia at 0xf7080000 irq 17
# 每一个声卡有一个card number和一个device number，可以用下面命令显示出来
$ aplay -l
> card 0: PCH [HDA Intel PCH], device 0: ALC887-VD Analog [ALC887-VD Analog]
Subdevices: 1/1
Subdevice #0: subdevice #0
card 0: PCH [HDA Intel PCH], device 1: ALC887-VD Digital [ALC887-VD Digital]
Subdevices: 1/1
Subdevice #0: subdevice #0
card 1: NVidia [HDA NVidia], device 3: HDMI 0 [HDMI 0]
Subdevices: 1/1
Subdevice #0: subdevice #0
card 1: NVidia [HDA NVidia], device 7: HDMI 1 [HDMI 1]
Subdevices: 1/1
Subdevice #0: subdevice #0
# 录音
$ arecord -D "plughw:0,0" -f S16_LE -r 16000 -d 5 -t wav file.wav
# -D 选择设备 试过hw:1,0 hw:0,2 只有hw:0,0可以录音
# -f 录音格式 S16_LE代表有符号16位小端序
# -r 采样率
# -t 录音时长
# file.wav 文件名
# 不添加plug会有警示,因为是外置声卡
Warning: rate is not accurate (requested = 16000Hz, got = 44100Hz) please, try the plug plugin
# 验证录音
$ aplay file.wav
一般alsa设置了一个defaults设备，音频播放软件默认使用defaults设备输出声音。defaults设备定义在alsa.conf中，内容如下：
[plain]
#
# defaults
#
# show all name hints also for definitions without hint {} section
defaults.namehint.showall off
# show just basic name hints
defaults.namehint.basic on
# show extended name hints
defaults.namehint.extended off
#
defaults.ctl.card 0
defaults.pcm.card 0
defaults.pcm.device 0
defaults.pcm.subdevice -1
……
defaults会默认匹配card number和device number比较小的声卡。
如果要修改，则修改/etc/asound.conf或~/.asoundrc。比如我要把defaults匹配到card 1,device 0上，则添加一下几行：
[plain]
$ sudo vim /etc/asound.conf
defaults.pcm.card 1
defaults.pcm.device 3
defaults.ctl.card 1
c++录音
https://github.com/aaronaanderson/ofxPortSF
有些可能记录时忘记记录获取信息的网站地址，有不当之处请指正~~
（若非特别声明，文章均为Vanessa的个人笔记，转载请注明出处。文章如有侵权内容，请联系我，我会及时删除）