Python 语音识别（pywin32）

看了一篇博客，写了几行代码，就实现了基本的语音识别，主要原理就是用pywin32库，中间也有一些小插曲，比如语音识别程序在后台没有关闭，导致出现了一个错误，

Traceback (most recent call last):
  File "E:\Python37\lib\site-packages\win32com\client\dynamic.py", line 89, in _GetGoodDispatch
    IDispatch = pythoncom.connect(IDispatch)
pywintypes.com_error: (-2147221021, '操作无法使用', None, None)

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "E:/Siri/Siri.py", line 3, in <module>
    import speech
  File "E:\Python37\lib\site-packages\speech.py", line 66, in <module>
    _recognizer = win32com.client.Dispatch("SAPI.SpSharedRecognizer")
  File "E:\Python37\lib\site-packages\win32com\client\__init__.py", line 95, in Dispatch
    dispatch, userName = dynamic._GetGoodDispatchAndUserName(dispatch,userName,clsctx)
  File "E:\Python37\lib\site-packages\win32com\client\dynamic.py", line 114, in _GetGoodDispatchAndUserName
    return (_GetGoodDispatch(IDispatch, clsctx), userName)
  File "E:\Python37\lib\site-packages\win32com\client\dynamic.py", line 91, in _GetGoodDispatch
    IDispatch = pythoncom.CoCreateInstance(IDispatch, None, clsctx, pythoncom.IID_IDispatch)
pywintypes.com_error: (-2147200911, 'OLE error 0x80045071', None, None)

，我上网查了这个错误，差点让我安装一个32位的python，后来打开任务管理器，将任务进程结束再次运行，成功。