import win32com.client #载入SAPI语音识别转换
from win32com.client import constants
import pythoncom #主要应用于python调用com接口

speaker=win32com.client.Dispatch('SAPI.SPVOICE')
speaker.Speak('开启微软语音接口')
speaker.Speak('Microsoft Speech API Initialized.')#英文语音合成

win32com.client.Dispatch('SAPI.SpSharedRecognizer')

import win32com.client #载入SAPI语音识别转换
from win32com.client import constants
import pythoncom #主要应用于python调用com接口
#定义语音识别对象并开启SAPI语音识别引擎
speaker=win32com.client.Dispatch('SAPI.SPVOICE')
#定义一个语音识别类
class SpeechRecognition:
    #用传入的单词列表初始化语音识别
    def __init__(self,wordsToAdd):
        #启动TTS
        self.speaker=win32com.client.Dispatch('SAPI.SpVoice')
        #启动语音识别引擎，首先创建一个侦听器
        self.listener=win32com.client.Dispatch("SAPI.SpSharedRecognizer")
        #创建语音识别上下文
        self.context=self.listener.CreateRecoContext()
        #不允许自由识别单词————仅限命令和控制识别语法中的单词
        self.grammar=self.context.CreateGrammar()
        #为语法创建一个新规则，即顶级规则和动态规则
        self.grammar.DictationSetState(0)
        self.wordsRule=self.grammar.Rules.Add("wordsRule",constants.SRATopLevel+\
                                              constants.SRADynamic,0)
        #清除规则
        self.wordsRule.Clear()
        #浏览单词列表
        [self.wordsRule.InitialState.AddWordTransition(None,word) for word in wordsToAdd]
        #将设置好的wordsrule规则激活
        self.grammar.Rules.Commit()
        self.grammar.CmdSetRuleState("wordsRule",1)
        #提交对语法规则的更改
        self.grammar.Rules.Commit()
        #添加一个事件处理程序
        self.eventHandler=ContextEvents(self.context)
        #设置一个语音提示
        self.say("Successfully Started.")
    #定义一个函数进行TTS语音输出
    def say(self,phrase):
        self.speaker.Speak(phrase)
#处理语音对象引发的事件的回调类
class ContextEvents(win32com.client.getevents("SAPI.SpSharedRecoContext")):
    def OnRecognition(self,StreamNumber,StreamPosition,RecognitionType,Result):
        newResult=win32com.client.Dispatch(Result)
        print("You just said:",newResult.PhraseInfo.GetText())
        speechstr=newResult.PhraseInfo.GetText()
        #定义语音识别关键词及其对应的语音输出信息
        if speechstr=="你好":
            speaker.Speak("How are you doing?")
        elif speechstr=="测试":
            speaker.Speak("This is a testing program for speech recognition.")
        elif speechstr=="欢迎你":
            speaker.Speak("You are welcome to be here")
        elif speechstr=='新年快乐':
            speaker.Speak("Happy New Year in 2019")
        elif speechstr=="作者":
            speaker.Speak("心随而动")
        else:
            pass
if __name__=='__main__':
    speaker.Speak("语音识别系统开启")
    wordsToAdd=['你好','测试','欢迎你','新年快乐','作者']
    speechReco=SpeechRecognition(wordsToAdd)
    while True:
        pythoncom.PumpWaitingMessages() #检查是否有事件在等待，然后再调用适当的回调

pip install speech

#speech
import speech
while True:
    phrase=speech.input() #循环等待语音输入
    speech.say("You said %s" %phrase)
    if phrase=='turn off':
        break
 如果你使用的是python3版本，那么下载后的speech是不能直接使用，需要你将代码改变一下。打开speech.py文件，

#语音控制计算机
import sys #用于windows系统操作
import speech
import webbrowser #用于打开指定网络连接
import os
def callback(phr):
    if phr=="关闭语音识别":
        speech.say("Goodbye,Speech Recognition system is closing.") #播放提示语音
        speech.stoplistening() #关闭SAPI
        sys.exit() #推出程序
    elif phr=="播放电影":
        speech.say("I am prepaing the movie for you.") #播放提示语音
        webbrowser.open_new('https://www.youku.com/')
    elif phr=="看新闻":
        speech.say("I want to know what the world is going on")
        webbrowser.open_new("https://www.baidu.com/")
    elif phr=="打开控制台":
        speech.say("打开CMD")
        os.popen("C:\\Windows\\System32\\cmd.exe")
while True:
    phr=speech.input() #语音输入
    speech.say("You said is %s"%phr)
    callback(phr) #调用函数

params
signal:
the audio signal from which to compute features.Should be an N1 array
用来计算梅尔频率倒谱系数特性的音频信号。是一个N1的数组
samplerate:
the samplerate of the signal we are working with.
音频信号的采样率
winlen:
the length of the analysis window in seconds. Default is 0.025s (25 milliseconds)
分析窗口的长度，以秒为单位。默认值为0.025s(25毫秒)（ps: 短时傅里叶变换的窗口长度）
winstep:
the step between successive windows in seconds. Default is 0.01s (10 milliseconds)
连续窗口之间的步长，以秒为单位。默认值为0.01s(10毫秒)
numcep:
the number of cepstrum to return, default 13
返回倒谱的数量，默认为13
nfilt:
the number of filters in the filterbank, default 26.
滤波器组中的过滤器数量，默认为26个。
nfft:
the FFT size. Default is 512.
FFT大小。默认是512。
lowfreq:
lowest band edge of mel filters. In Hz, default is 0.
梅尔滤波器的最低频带边缘。在频率(HZ)中，默认值为0。
highfreq:
highest band edge of mel filters. In Hz, default is samplerate/2
梅尔过滤器的最高频带边缘。在频率(HZ)中，默认值为1/2倍音频信号采样率
preemph:
apply preemphasis filter with preemph as coefficient. 0 is no filter. Default is 0.97.
采用preemph为系数的预加重滤波器。0不是过滤器。默认是0.97。
ceplifter:
apply a lifter to final cepstral coefficients. 0 is no lifter. Default is 22.
将一个lifter应用到最终倒谱系数。0不是lifter。默认是22。
appendEnergy:
if this is true, the zeroth cepstral coefficient is replaced with the log of the total frame energy.
如果这个参数的值是True，第0阶倒谱系数被替换为总帧能量的对数。
winfunc:
the analysis window to apply to each frame. By default no window is applied. You can use numpy window functions here e.g. winfunc=numpy.hamming
分析窗口应用于每一帧。默认情况下不应用任何窗口。您可以在这里使用numpy窗口函数，例如winfunc=numpy.hamming(汉明窗)c

from python_speech_features import mfcc
from python_speech_features import logfbank
import scipy.io.wavfile as wav #用于语音文件载入
(rate,sig)=wav.read("file.wav") #载入音频文件，大家自己使用自己的音频文件，这里博主随便写了一个
mfcc_feat=mfcc(sig,rate)
fbank_feat=logfbank(sig,rate)    #计算滤波器组能量特征的对数
#输出结果
print("MFCC Features:")
print(mfcc_feat[1:3,:])
print("LogFBank Features:")
print(fbank_feat[1:3,:])
SpeechRecognition库工具
SpeechRecognition是一个用于语音识别的Python库，他同时支持python2和python3联机或离线的多个引擎和API。

import speech_recognition as sr#导入SpeechRecognition库
#从话筒获取语音识别的音频源
r=sr.Recognizer()
with sr.Microphone() as source:
    print("Say something")
    audio=r.listen(source)
#利用Cmu Sphinx进行语音识别
try:
    print("Sphinx thinks you said"+r.recognize_sphinx(audio))
except sr.UnknownValueError:
    print("Sphinx could not undestand audio")
except sr.RequestError as e:
    print("Sphinx error;{0}".format(e))
#利用Google Speech Recognition进行语音识别
try:
    '''
    处于测试的目的，我们使用默认的API密钥，若要使用其他的API密钥，建议使用
    r.recognize_google(audio,key="GOOGLE_SPEECH_RECOGNITION_API_KEY"),而不是用r.recognize_google(audio)
    '''
    print("Google Speech Recognition thinks you said"+r.recognize_google(audio))
except sr.UnknownValueError:
    print("Google Speech Recognition could not wnderstand audio")
except sr.RequestError as e:
    print("Could not equest results from Google Speech Recognition service;{0}".format(e))
GOOGLE_CLOUD_SPEECH_CREDENTIALS=r"""INSERT THE CONTENTS OF THE GOOGLE CLOUD SPEECH JSON CREDENTIALS FILE HERE"""
try:
    print("Google cloud speech thinks you said"+r.recognize_google_cloud(audio,credentials_json=GOOGLE_CLOUD_SPEECH_CREDENTIALS))
except sr.UnknownValueError:
    print("Google Cloud Speech could not understand audio")
except sr.RequestError as e:
    print("Could not request results from google cloud Speech service;{0}".format(e))
#利用Microsoft Bing voice Recognition进行语音识别
BING_KEY="INSERT BING API KEY HERE"
try:
    print("Microsoft Bing Voice Recognition thinks you said"+r.recognize_bing(audio,key=BING_KEY))
except sr.UnknownValueError:
    print("Microsoft Bing Voice Recognition could not understand audio")
except sr.RequestError as e:
    print("Could not request results from Microsoft Bing Voice Recognition service;{0}".format(e))
#利用Houndify进行语音识别。Hounddify客户端ID是Base64编码字符串
HOUNDIFY_CLIENT_ID="INSERT HOUNDIFY CLIENT ID HERE"
#加密客户端密钥是Base64编码字符串
HOUNDIFY_CLIENT_KEY="INSERT HOUNDIFY CLIENT KEY HERE"
try:
    print("Houndify thinks you said"+r.recognize_houndify(audio,client_id=HOUNDIFY_CLIENT_ID,client_key=HOUNDIFY_CLIENT_KEY))
except sr.UnknownValueError:
    print("Houndify could not understand audio")
except sr.RequestError as e:
    print("Could not request results from Houndify service;{0}".format(e))
    '''
    利用IBM Speech to Text进行语音识别
    '''
    IBM_USERNAME="INSERT IBM SPEECH TO TEXT USERNAME HERE"
    #IBM Speech to Text密码是混合大小，小写字母和数字的字符串
    IBM_PASSWORD="INSERT IBM SPEECH TO TEXT PASSWORD HERE"
    try:
        print("IBM Speech to Text thinks you said"+r.recognize_ibm(audio,username=IBM_USERNAME,password=IBM_PASSWORD))
    except sr.UnknownValueError:
        print("IBM Speechh to Text could not understand audio")
    except sr.RequestError as e:
        print("Could not request results from IBM Speech to Text service;{0}".format(e))

from os import path
AUDIO_FILE=path.join(path.dirname(path.realpath(__file__)),"English.wav") #导入音频文件，大家自己更改路径
r=sr.Recognizer()
with sr.AudioFile(AUDIO_FILE) as source:
    audio=r.record(source)
除了使用音频文件，我么也可以使用话筒来获取语音并存储为不同格式的文件
import speech_recognition as sr
r=sr.Recognizer()#从话筒获取语音
with sr.Microphone() as source:
    print("Say something!")
    audio=r.listen(source)
#将音频写入.raw文件
with open('microphone-results.raw','wb') as f:
    f.write(audio.get_raw_data())
#写入.wav文件
with open('microphone-results1.wav','wb') as f:
    f.write(audio.get_wav_data())
#写入.aiff文件
with open('microphone-results2.aiff','wb') as f:
    f.write(audio.get_aiff_data())
#写入.flac文件
with open('microphone-results.flac','wb') as f:
    f.write(audio.get_flac_data())

r=sr.Recognizer()
with sr.Microphone()as source:
    #倾听1秒以内能量校准阈值
    r.adjust_for_ambient_noise(source)
    print("Say something!")
    audio=r.listen(source)