diff --git a/speech_features/__init__.py b/speech_features/__init__.py index c3144f0..f5ed512 100644 --- a/speech_features/__init__.py +++ b/speech_features/__init__.py @@ -18,4 +18,9 @@ # along with ASRT. If not, see . # ============================================================================ -from .speech_features import * \ No newline at end of file +''' +@author: nl8590687 +ASRT语音识别声学特征模块 +''' + +from .speech_features import * diff --git a/speech_features/base.py b/speech_features/base.py index e1988a0..c439f8c 100644 --- a/speech_features/base.py +++ b/speech_features/base.py @@ -20,6 +20,12 @@ # calculate filterbank features. Provides e.g. fbank and mfcc features for use in ASR applications # Author: James Lyons 2012 + +''' +@author: nl8590687 +ASRT语音识别声学特征基础库模块，一些基础函数实现 +''' + from __future__ import division import numpy from scipy.fftpack import dct @@ -29,7 +35,7 @@ from .sigproc import preemphasis, framesig, powspec def calculate_nfft(samplerate, winlen): """Calculates the FFT size as a power of two greater than or equal to the number of samples in a single window length. - + Having an FFT less than the window length loses precision by dropping many of the samples; a longer FFT than the window allows zero-padding of the FFT buffer which is neutral in terms of frequency domain conversion. @@ -218,4 +224,4 @@ def delta(feat, N): padded = numpy.pad(feat, ((N, N), (0, 0)), mode='edge') # padded version of feat for t in range(NUMFRAMES): delta_feat[t] = numpy.dot(numpy.arange(-N, N+1), padded[t : t+2*N+1]) / denominator # [t : t+2*N+1] == [(N+t)-N : (N+t)+N+1] - return delta_feat \ No newline at end of file + return delta_feat diff --git a/speech_features/sigproc.py b/speech_features/sigproc.py index 1161740..da17a79 100644 --- a/speech_features/sigproc.py +++ b/speech_features/sigproc.py @@ -20,10 +20,16 @@ # This file includes routines for basic signal processing including framing and computing power spectra. # Author: James Lyons 2012 + +''' +@author: nl8590687 +ASRT语音识别声学特征计算的信号处理计算的函数库 +''' + import decimal -import numpy -import math import logging +import math +import numpy def round_half_up(number): @@ -135,7 +141,7 @@ def logpowspec(frames, NFFT, norm=1): :param norm: If norm=1, the log power spectrum is normalised so that the max value (across all frames) is 0. :returns: If frames is an NxD matrix, output will be Nx(NFFT/2+1). Each row will be the log power spectrum of the corresponding frame. """ - ps = powspec(frames, NFFT); + ps = powspec(frames, NFFT) ps[ps <= 1e-30] = 1e-30 lps = 10 * numpy.log10(ps) if norm: @@ -150,4 +156,4 @@ def preemphasis(signal, coeff=0.95): :param coeff: The preemphasis coefficient. 0 is no filter, default is 0.95. :returns: the filtered signal. """ - return numpy.append(signal[0], signal[1:] - coeff * signal[:-1]) \ No newline at end of file + return numpy.append(signal[0], signal[1:] - coeff * signal[:-1]) diff --git a/speech_features/speech_features.py b/speech_features/speech_features.py index 684cd94..2f74be1 100644 --- a/speech_features/speech_features.py +++ b/speech_features/speech_features.py @@ -20,13 +20,11 @@ """ @author: nl8590687 -ASRT语音识别内置声学特征提取模块 - +ASRT语音识别内置声学特征提取模块，定义了几个常用的声学特征类 """ import random import numpy as np -import math from scipy.fftpack import fft from .base import mfcc, delta, logfbank @@ -38,10 +36,12 @@ class SpeechFeatureMeta(): ''' def __init__(self, framesamplerate = 16000): self.framesamplerate = framesamplerate - pass - + def run(self, wavsignal, fs = 16000): - raise NotImplementedError('[ASRT] Get speech feature function is not implemented. Please define "a run method"') + ''' + run method + ''' + raise NotImplementedError('[ASRT] `run()` method is not implemented.') class MFCC(SpeechFeatureMeta): ''' @@ -56,7 +56,12 @@ class MFCC(SpeechFeatureMeta): :param nfilt: the number of filters in the filterbank, default 26. :param preemph: apply preemphasis filter with preemph as coefficient. 0 is no filter. Default is 0.97. ''' - def __init__(self, framesamplerate = 16000, winlen=0.025, winstep=0.01, numcep=13, nfilt=26, preemph=0.97): + def __init__(self, framesamplerate = 16000, + winlen=0.025, + winstep=0.01, + numcep=13, + nfilt=26, + preemph=0.97): self.framesamplerate = framesamplerate self.winlen = winlen self.winstep = winstep @@ -64,7 +69,7 @@ class MFCC(SpeechFeatureMeta): self.nfilt = nfilt self.preemph = preemph super().__init__(framesamplerate) - + def run(self, wavsignal, fs = 16000): ''' 计算mfcc声学特征，包含静态特征、一阶差分和二阶差分 @@ -73,7 +78,8 @@ class MFCC(SpeechFeatureMeta): ''' wavsignal = np.array(wavsignal, dtype=np.float) # 获取输入特征 - feat_mfcc=mfcc(wavsignal[0], samplerate=self.framesamplerate, winlen=self.winlen, winstep=self.winstep, numcep=self.numcep, nfilt=self.nfilt, preemph=self.preemph) + feat_mfcc=mfcc(wavsignal[0], samplerate=self.framesamplerate, winlen=self.winlen, + winstep=self.winstep, numcep=self.numcep, nfilt=self.nfilt, preemph=self.preemph) feat_mfcc_d=delta(feat_mfcc, 2) feat_mfcc_dd=delta(feat_mfcc_d, 2) # 返回值分别是mfcc特征向量的矩阵及其一阶差分和二阶差分矩阵 @@ -86,7 +92,7 @@ class Logfbank(SpeechFeatureMeta): ''' def __init__(self, framesamplerate = 16000): super().__init__(framesamplerate) - + def run(self, wavsignal, fs = 16000): wavsignal = np.array(wavsignal, dtype=np.float) # 获取输入特征 @@ -94,9 +100,13 @@ class Logfbank(SpeechFeatureMeta): return wav_feature class Spectrogram(SpeechFeatureMeta): + ''' + ASRT语音识别内置的语谱图声学特征提取类 + ''' def __init__(self, framesamplerate = 16000, timewindow = 25, timeshift = 10): self.time_window = timewindow self.window_length = int(framesamplerate / 1000 * self.time_window) # 计算窗长度的公式，目前全部为400固定值 + self.timeshift = timeshift ''' # 保留将来用于不同采样频率 @@ -107,47 +117,45 @@ class Spectrogram(SpeechFeatureMeta): self.x=np.linspace(0, 400 - 1, 400, dtype = np.int64) self.w = 0.54 - 0.46 * np.cos(2 * np.pi * (self.x) / (400 - 1) ) # 汉明窗 super().__init__(framesamplerate) - + def run(self, wavsignal, fs = 16000): - if(16000 != fs): + if fs != 16000: raise ValueError('[Error] ASRT currently only supports wav audio files with a sampling rate of 16000 Hz, but this audio is ' + str(fs) + ' Hz. ') - + # wav波形 加时间窗以及时移10ms time_window = 25 # 单位ms window_length = int(fs / 1000 * time_window) # 计算窗长度的公式，目前全部为400固定值 - + wav_arr = np.array(wavsignal) #wav_length = len(wavsignal[0]) - wav_length = wav_arr.shape[1] - + #wav_length = wav_arr.shape[1] + range0_end = int(len(wavsignal[0])/fs*1000 - time_window) // 10 + 1 # 计算循环终止的位置，也就是最终生成的窗数 data_input = np.zeros((range0_end, window_length // 2), dtype = np.float) # 用于存放最终的频率特征数据 data_line = np.zeros((1, window_length), dtype = np.float) - + for i in range(0, range0_end): p_start = i * 160 p_end = p_start + 400 - + data_line = wav_arr[0, p_start:p_end] - data_line = data_line * self.w # 加窗 - - #data_line = np.abs(fft(data_line)) / wav_length data_line = np.abs(fft(data_line)) - + data_input[i]=data_line[0: window_length // 2] # 设置为400除以2的值（即200）是取一半数据，因为是对称的 - + #print(data_input.shape) data_input = np.log(data_input + 1) return data_input class SpecAugment(SpeechFeatureMeta): ''' - 复现谷歌SpecAugment数据增强算法 + 复现谷歌SpecAugment数据增强特征算法，基于Spectrogram语谱图基础特征 ''' def __init__(self, framesamplerate = 16000, timewindow = 25, timeshift = 10): self.time_window = timewindow self.window_length = int(framesamplerate / 1000 * self.time_window) # 计算窗长度的公式，目前全部为400固定值 + self.timeshift = timeshift ''' # 保留将来用于不同采样频率 @@ -158,36 +166,33 @@ class SpecAugment(SpeechFeatureMeta): self.x=np.linspace(0, 400 - 1, 400, dtype = np.int64) self.w = 0.54 - 0.46 * np.cos(2 * np.pi * (self.x) / (400 - 1) ) # 汉明窗 super().__init__(framesamplerate) - + def run(self, wavsignal, fs = 16000): - if(16000 != fs): + if fs != 16000: raise ValueError('[Error] ASRT currently only supports wav audio files with a sampling rate of 16000 Hz, but this audio is ' + str(fs) + ' Hz. ') - + # wav波形 加时间窗以及时移10ms time_window = 25 # 单位ms window_length = int(fs / 1000 * time_window) # 计算窗长度的公式，目前全部为400固定值 - + wav_arr = np.array(wavsignal) #wav_length = len(wavsignal[0]) - wav_length = wav_arr.shape[1] - + #wav_length = wav_arr.shape[1] + range0_end = int(len(wavsignal[0])/fs*1000 - time_window) // 10 + 1 # 计算循环终止的位置，也就是最终生成的窗数 data_input = np.zeros((range0_end, window_length // 2), dtype = np.float) # 用于存放最终的频率特征数据 data_line = np.zeros((1, window_length), dtype = np.float) - + for i in range(0, range0_end): p_start = i * 160 p_end = p_start + 400 - + data_line = wav_arr[0, p_start:p_end] - data_line = data_line * self.w # 加窗 - - #data_line = np.abs(fft(data_line)) / wav_length data_line = np.abs(fft(data_line)) - + data_input[i]=data_line[0: window_length // 2] # 设置为400除以2的值（即200）是取一半数据，因为是对称的 - + #print(data_input.shape) data_input = np.log(data_input + 1) @@ -199,17 +204,13 @@ class SpecAugment(SpeechFeatureMeta): v_start = random.randint(1,data_input.shape[1]) v_width = random.randint(1,100) - if(mode <= 60): # 正常特征 60% + if mode <= 60: # 正常特征 60% pass - elif(mode > 60 and mode <=75): # 横向遮盖 15% + elif 60 < mode <=75: # 横向遮盖 15% data_input[h_start:h_start+h_width,:] = 0 - pass - elif(mode > 75 and mode <= 90): # 纵向遮盖 15% + elif 75 < mode <= 90: # 纵向遮盖 15% data_input[:,v_start:v_start+v_width] = 0 - pass else: # 两种遮盖叠加 10% data_input[h_start:h_start+h_width,:v_start:v_start+v_width] = 0 - pass - return data_input