演讲文本 [英] Speech to Text

问题描述

我们正在考虑使用Speech to Text来满足我们的转录需求，数据本质上是非常敏感的，所以我想知道你们是否使用以任何方式发送到服务器的任何音频。转录它并将价值归还给我们？我还要
想要确保转录文本在处理完毕后被处理掉并且不存储在任何地方。

We're considering Speech to Text for our transcription needs, the data is pretty sensitive in nature so I would like to know if you guys use any of the audio sent over to your servers in any way other than to transcribe it and return the value to us? I also want to make sure that the transcribed text is disposed of and not stored anywhere after the call is done.

谢谢。

推荐答案

抱歉，我不知道项目oxford spreech API的完整隐私政策，但在"开发者行为准则"中说：

Sorry, I don't know the complete privacy policy for the project oxford spreech API, but in the Code of Conduct for Developers it says:

• 由于您使用的是Project Oxford API，您还需要在隐私政策中加入此模糊：
  
  
  
  
  • 图片，语音和/或视频理解功能
    <<您的app name here>> app使用Microsoft Project Oxford API。 Microsoft将接收您上传的图像，音频，视频和其他数据（通过此应用程序），以便进行服务改进。要向Microsoft报告滥用Project Oxford API
    ，请访问Microsoft Project Oxford网站 www.projectoxford.ai ，
    并使用底部的"举报滥用行为"链接要联系Microsoft的页面。有关Microsoft隐私政策的更多信息，请参阅此处的隐私声明：
    http://go.microsoft.com/fwlink/?LinkId=521839 。
  • Since you are using the Project Oxford APIs, you'll also need to include this blurb in your privacy policy:
    • The image, voice, and/or video understanding capabilities of <<your app name here>> app use the Microsoft Project Oxford APIs. Microsoft will receive the images, audio, video, and other data that you upload (via this app) for service improvement purposes. To report abuse of the Project Oxford APIs to Microsoft, please visit the Microsoft Project Oxford website at www.projectoxford.ai, and use the "Report Abuse" link at the bottom of the page to contact Microsoft. For more information about Microsoft privacy policies please see their privacy statement here: http://go.microsoft.com/fwlink/?LinkId=521839.

    http://research.microsoft.com/en-us/UM/legal/Dev_Code_of_Conduct_ProjOxfordAPIs-20151106.htm

    您可以在此处找到完整的服务协议：

    You can find the full services agreement here:

    http://research.microsoft.com/en-us /UM/legal/Project_Oxford_Online_Services_Agreement_20151130.html

    但如果您正在构建Windows Universal App您还可以使用语音识别命名空间中的语音识别器构建：

    But if you a building a Windows Universal App you can also use the build in Speech Recognizer from the Speech Recognition Namespace:

    https://msdn.microsoft.com/de-de/library/windows.media.speechrecognition.aspx

    这将允许你做语音到本地文本。

    This will allow you to do speech to Text locally.

    最好，

    Malte

    
    

    这篇关于演讲文本的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！