语音识别-NSSpeechRecognizer或其他？

Question

我已阅读NSSpeechRecognizer的Apple文档，我仍然致电openears。

苹果这样说：

The NSSpeechRecognizer class is the Cocoa interface to Speech Recognition on OS X. Speech Recognition is architected as a “command and control” voice recognition system. It uses a finite state grammar and listens for phrases in that grammar. When it recognizes a phrase, it notifies the client process. This architecture is different from that used to support dictation.

我无法理解其实际意义。 当按钮被按下时，我们需要在iPhone上听用户，而不是识别所说的单词和数字。 识别应该与用户说话一样，而不是在停止时，它返回单词/单词。 我们不希望它停在中间，如：3百 - >不会返回3 ..但等到沉默。

OpenEar更好还是Apple原生的？

谢谢。