ReSpeaker Mic Array v2.0 快速上手
参考自:https://wiki.seeedstudio.com/ReSpeaker_Mic_Array_v2.0/#doa-direction-of-arrival
https://wiki.seeedstudio.com/cn/ReSpeaker_Mic_Array_v2.0/
我的环境
树莓派3
作用
该功能类似天猫精灵,即在设备播放过程中检测唤醒词用于打断或重定向任务。
即让智能设备在"说"的时候同时还可以"听" ,人类是有意识的,所以自己说的话和外界的声音很容易分辨,但是机器不行。
根据介绍,该设备中有一些算法 (DSP 算法,包括声学回声消除 (AEC),波束成形,去混响,噪声抑制和增益控制。)但是最有意思的是声源定位,这个很好玩。
设备
该设备主要三个类型,贵一点的是集成开发环境、便宜点的是插在树莓派上使用的,但是gpio都被占用这个过分了。我用的是usb的,因为linux 免驱动,这个还是挺方便的。
上手
该设备主要有两种固件,主要区别就是通道数不同,上手第一件事就是更新固件,因为6通道的版本功能强了不少。
通道 0 : ASR 处理音频,通道 1 : mic1 原始数据,通道 2 : mic2 原始数据,通道 3 : mic3 原始数据,通道 4 : mic4 原始数据,通道 5 : 合并播放
0通道就是我们需要的,该设备内置的算法会放大你说话的音量、降低背景音量,但是又不能完全去除背景音
接下来
git clone https://github.com/respeaker/usb_4_mic_array.git
cd usb_4_mic_array
python tuning.py -p
使用tuning.py可以设置设备内部参数具体如下:
#自适应回声抵消器更新抑制 0 自适应已启用 1 冻结自适应,仅限过滤器
'AECFREEZEONOFF': (18, 7, 'int', 1, 0, 'rw', 'Adaptive Echo Canceler updates inhibit.', '0 = Adaptation enabled', '1 = Freeze adaptation, filter only'),
#AEC滤波器系数范数的极限
'AECNORM': (18, 19, 'float', 16, 0.25, 'rw', 'Limit on norm of AEC filter coefficients'),
#AEC路径更改检测 0 false 1 true
'AECPATHCHANGE': (18, 25, 'int', 1, 0, 'ro', 'AEC Path Change Detection.', '0 = false (no path change detected)', '1 = true (path change detected)'),
#当前RT60估计值(秒)
'RT60': (18, 26, 'float', 0.9, 0.25, 'ro', 'Current RT60 estimate in seconds'),
#麦克风信号的高通滤波器
'HPFONOFF': (18, 27, 'int', 3, 0, 'rw', 'High-pass Filter on microphone signals.', '0 = OFF', '1 = ON - 70 Hz cut-off', '2 = ON - 125 Hz cut-off', '3 = ON - 180 Hz cut-off'),
#RT60 AES估算
'RT60ONOFF': (18, 28, 'int', 1, 0, 'rw', 'RT60 Estimation for AES. 0 = OFF 1 = ON'),
#AEC中信号检测的阈值
'AECSILENCELEVEL': (18, 30, 'float', 1, 1e-09, 'rw', 'Threshold for signal detection in AEC [-inf .. 0] dBov (Default: -80dBov = 10log10(1x10-8))'),
#AEC远端静音检测状态 0检测信号 1 静音检测
'AECSILENCEMODE': (18, 31, 'int', 1, 0, 'ro', 'AEC far-end silence detection status. ', '0 = false (signal detected) ', '1 = true (silence detected)'),
#自动增益控制
'AGCONOFF': (19, 0, 'int', 1, 0, 'rw', 'Automatic Gain Control. ', '0 = OFF ', '1 = ON'),
#最大AGC增益因数
'AGCMAXGAIN': (19, 1, 'float', 1000, 1, 'rw', 'Maximum AGC gain factor. ', '[0 .. 60] dB (default 30dB = 20log10(31.6))'),
#输出信号的目标功率电平
'AGCDESIREDLEVEL': (19, 2, 'float', 0.99, 1e-08, 'rw', 'Target power level of the output signal. ', '[−inf .. 0] dBov (default: −23dBov = 10log10(0.005))'),
#当前AGC增益因数
'AGCGAIN': (19, 3, 'float', 1000, 1, 'rw', 'Current AGC gain factor. ', '[0 .. 60] dB (default: 0.0dB = 20log10(1.0))'),
#上升/下降时间常数,以秒为单位。
'AGCTIME': (19, 4, 'float', 1, 0.1, 'rw', 'Ramps-up / down time-constant in seconds.'),
#舒适性噪音插入
'CNIONOFF': (19, 5, 'int', 1, 0, 'rw', 'Comfort Noise Insertion.', '0 = OFF', '1 = ON'),
#自适应波束形成器更新
'FREEZEONOFF': (19, 6, 'int', 1, 0, 'rw', 'Adaptive beamformer updates.', '0 = Adaptation enabled', '1 = Freeze adaptation, filter only'),
#静止噪声抑制
'STATNOISEONOFF': (19, 8, 'int', 1, 0, 'rw', 'Stationary noise suppression.', '0 = OFF', '1 = ON'),
#平稳噪声的过减因子。最小值。。最大衰减
'GAMMA_NS': (19, 9, 'float', 3, 0, 'rw', 'Over-subtraction factor of stationary noise. min .. max attenuation'),
#固定噪声抑制增益下限
'MIN_NS': (19, 10, 'float', 1, 0, 'rw', 'Gain-floor for stationary noise suppression.', '[−inf .. 0] dB (default: −16dB = 20log10(0.15))'),
#非平稳噪声抑制
'NONSTATNOISEONOFF': (19, 11, 'int', 1, 0, 'rw', 'Non-stationary noise suppression.', '0 = OFF', '1 = ON'),
#非平稳噪声的过减因子。最小值。。最大衰减
'GAMMA_NN': (19, 12, 'float', 3, 0, 'rw', 'Over-subtraction factor of non- stationary noise. min .. max attenuation'),
#非平稳噪声抑制的增益下限
'MIN_NN': (19, 13, 'float', 1, 0, 'rw', 'Gain-floor for non-stationary noise suppression.', '[−inf .. 0] dB (default: −10dB = 20log10(0.3))'),
#回声抑制
'ECHOONOFF': (19, 14, 'int', 1, 0, 'rw', 'Echo suppression.', '0 = OFF', '1 = ON'),
#回波的过减因子(直接分量和早期分量)。最小值。。最大衰减
'GAMMA_E': (19, 15, 'float', 3, 0, 'rw', 'Over-subtraction factor of echo (direct and early components). min .. max attenuation'),
#回波(尾分量)的过减因子。最小值。。最大衰减
'GAMMA_ETAIL': (19, 16, 'float', 3, 0, 'rw', 'Over-subtraction factor of echo (tail components). min .. max attenuation'),
#非线性回波的过减因子。最小值。。最大衰减
'GAMMA_ENL': (19, 17, 'float', 5, 0, 'rw', 'Over-subtraction factor of non-linear echo. min .. max attenuation'),
#非线性回波衰减。
'NLATTENONOFF': (19, 18, 'int', 1, 0, 'rw', 'Non-Linear echo attenuation.', '0 = OFF', '1 = ON'),
#非线性AEC训练模式。
'NLAEC_MODE': (19, 20, 'int', 2, 0, 'rw', 'Non-Linear AEC training mode.', '0 = OFF', '1 = ON - phase 1', '2 = ON - phase 2'),
#语音检测状态
'SPEECHDETECTED': (19, 22, 'int', 1, 0, 'ro', 'Speech detection status.', '0 = false (no speech detected)', '1 = true (speech detected)'),
#FSB更新决策
'FSBUPDATED': (19, 23, 'int', 1, 0, 'ro', 'FSB Update Decision.', '0 = false (FSB was not updated)', '1 = true (FSB was updated)'),
#FSB路径变化检测
'FSBPATHCHANGE': (19, 24, 'int', 1, 0, 'ro', 'FSB Path Change Detection.', '0 = false (no path change detected)', '1 = true (path change detected)'),
#瞬态回波抑制
'TRANSIENTONOFF': (19, 29, 'int', 1, 0, 'rw', 'Transient echo suppression.', '0 = OFF', '1 = ON'),
#VAD语音活动状态
'VOICEACTIVITY': (19, 32, 'int', 1, 0, 'ro', 'VAD voice activity status.', '0 = false (no voice activity)', '1 = true (voice activity)'),
#ASR的平稳噪声抑制
'STATNOISEONOFF_SR': (19, 33, 'int', 1, 0, 'rw', 'Stationary noise suppression for ASR.', '0 = OFF', '1 = ON'),
#ASR的非平稳噪声抑制
'NONSTATNOISEONOFF_SR': (19, 34, 'int', 1, 0, 'rw', 'Non-stationary noise suppression for ASR.', '0 = OFF', '1 = ON'),
#ASR平稳噪声的过减因子
'GAMMA_NS_SR': (19, 35, 'float', 3, 0, 'rw', 'Over-subtraction factor of stationary noise for ASR. ', '[0.0 .. 3.0] (default: 1.0)'),
#ASR非平稳噪声的过减因子。
'GAMMA_NN_SR': (19, 36, 'float', 3, 0, 'rw', 'Over-subtraction factor of non-stationary noise for ASR. ', '[0.0 .. 3.0] (default: 1.1)'),
#
'MIN_NS_SR': (19, 37, 'float', 1, 0, 'rw', 'Gain-floor for stationary noise suppression for ASR.', '[−inf .. 0] dB (default: −16dB = 20log10(0.15))'),
#ASR的固定噪声抑制增益下限。
'MIN_NN_SR': (19, 38, 'float', 1, 0, 'rw', 'Gain-floor for non-stationary noise suppression for ASR.', '[−inf .. 0] dB (default: −10dB = 20log10(0.3))'),
#设置语音活动检测的阈值。
'GAMMAVAD_SR': (19, 39, 'float', 1000, 0, 'rw', 'Set the threshold for voice activity detection.', '[−inf .. 60] dB (default: 3.5dB 20log10(1.5))'),
# 'KEYWORDDETECT': (20, 0, 'int', 1, 0, 'ro', 'Keyword detected. Current value so needs polling.'),
#方位角。当前值。方向取决于生成配置。
'DOAANGLE': (21, 0, 'int', 359, 0, 'ro', 'DOA angle. Current value. Orientation depends on build configuration.')
这些内部参数可以通过该指令来设置比如关闭自动增益 python tuning.py AGCONOFF 0
也可以使用代码读取其中的参数值
这段代码实际上读取的是VOICEACTIVITY这个参数值,该参数的阀值可以通过GAMMAVAD_SR来控制
需要注意的是这些个参数设置完了以后,重新启动即会回复为默认值。
重点是这些调整参数会直接影响到通道中的数据结果。
声源定位
读取的是DOAANGLE 也就是最后一个参数。
如果遇到usb.core.USBError: [Errno 13] Access denied
try:https://blog.csdn.net/weixin_43928944/article/details/109742040
pyaudio获取0通道数据
import pyaudio
import wave
import numpy as np
RESPEAKER_RATE = 16000
RESPEAKER_CHANNELS = 6 # change base on firmwares, 1_channel_firmware.bin as 1 or 6_channels_firmware.bin as 6
RESPEAKER_WIDTH = 2
# run getDeviceInfo.py to get index
RESPEAKER_INDEX = 0 # refer to input device id
CHUNK = 1024
RECORD_SECONDS = 7
WAVE_OUTPUT_FILENAME = "output.wav"
p = pyaudio.PyAudio()
stream = p.open(
rate=RESPEAKER_RATE,
format=p.get_format_from_width(RESPEAKER_WIDTH),
channels=RESPEAKER_CHANNELS,
input=True,
input_device_index=RESPEAKER_INDEX, )
print("* recording")
frames = []
for i in range(0, int(RESPEAKER_RATE / CHUNK * RECORD_SECONDS)):
data = stream.read(CHUNK)
# extract channel 0 data from 6 channels, if you want to extract channel 1, please change to [1::6]
a = np.frombuffer(data, dtype=np.int16)[0::6]
frames.append(a.tostring())
print("* done recording")
stream.stop_stream()
stream.close()
p.terminate()
wf = wave.open(WAVE_OUTPUT_FILENAME, 'wb')
wf.setnchannels(1)
print(p.get_sample_size(p.get_format_from_width(RESPEAKER_WIDTH)))
wf.setsampwidth(p.get_sample_size(p.get_format_from_width(RESPEAKER_WIDTH)))
wf.setframerate(RESPEAKER_RATE)
wf.writeframes(b''.join(frames))
wf.close()
device id可以用以下代码查看
import pyaudio
p = pyaudio.PyAudio()
info = p.get_host_api_info_by_index(0)
numdevices = info.get('deviceCount')
for i in range(0, numdevices):
if (p.get_device_info_by_host_api_device_index(0, i).get('maxInputChannels')) > 0:
print("Input Device id ", i, " - ", p.get_device_info_by_host_api_device_index(0, i).get('name'))
接下来聊聊free-avs(语音助手实现,自由对话)
主要代码在voice-engine中
git clone https://github.com/voice-engine/voice-engine
由于ReSpeaker使用来大量的三方开源库,环境搭建起来超级麻烦,这个跟它的价格形成了鲜明的对比,显然有些low
在voice-engine你可以看到比如回声处理和降噪都是用现成的
降噪
看一下 voice_engine中的栗子🌰
根据栗子可以看出实现的粗糙结构是这样的(画图忒麻烦,凑合看)
然后看看这个snowboy,它有个问题,就是官网训练的是特定某个人的音频信息,而通用模型…
使用的alexa_02092017.umdl通用模型
但是snowboy的通用模型没有中文的
而且
不太妙… 官方停止该服务后也没有提及之后怎么办,所以如果凉了还是要找替代方案。当然可以用开源的ASR自己训练模型,但是搞数据集是件挺麻烦的事情,负样本好解决,正样本就有些麻烦了。
写到这在写下去没意义了,此路不通的话打算去尝试一下使用Mycroft替代snowboy。