The Speech Platform Runtime 11 and the Speech Platform SDK 11 do not include Runtime Languages for speech recognition or for speech synthesis (TTS or text-to-speech). You must install them separately. A Runtime Language includes the language model, acoustic model, and other data necessary to provision a speech engine to perform speech recognition or TTS in a particular language. There are separate Runtime Languages for speech recognition or speech synthesis. The version of Runtime Languages that you download (for example, version 11.0) must match the version of the Speech Platform Runtime that you have installed. You can download Runtime Languages using this link.

上面描述的安裝步驟為：

1、卸載之前安裝過的任何版本的Speech Platform SDK 庫

2、確保已經(jīng)安裝了.net4.0

3、根據(jù)你是x86還是x64系統(tǒng)，下載對應(yīng)系統(tǒng)下的Speech Platform Runtime 11 和Speech Platform SDK 11

4、下載所需語言下的語音識別語言包和tts語言包，因為前兩個下載的庫里面并不包含運(yùn)行時的語言包，需要根據(jù)需要識別的語言種類或合成的語言種類去下載。

總結(jié)：

對于中文語音識別和語音合成，需要下載下面幾個安裝文件：

SpeechPlatformRuntime_x86.msi            (x86平臺下的運(yùn)行時庫)

MicrosoftSpeechPlatformSDK_x86.msi   (x86平臺下的sdk庫)

MSSpeech_SR_zh-CN_TELE.msi  (語音識別中文語言包)

MSSpeech_TTS_zh-CN_HuiHui.msi (語音合成中文語言包)