Microsoft’s Cognitive Services Comes to Azure Media Services in Preview

Microsoft has been developing a set of services for speech, face and emotion recognition.  Known previously as “Project Oxford”, these services have now been bundled as Microsoft Cognitive Services.

Microsoft Project Oxford

For example, the APIs can detect the two faces and the feeling of surprise in this photo.

Microsoft has now announced these services are coming to Azure Media Services, branded as “Azure Media Analytics”.  Azure Media Services is Microsoft’s video and audio encoding service that allows you to encode audio and video and have it served to a mass audience on a variety of formats from mobile phones to broadcast quality. 

A few scenarios for using these new services include:

  • Analysis of customer service or call center audio for emotional content
  • Face detection within security video
  • Analysis of body cams, dashcams, etc. for evidence for policing
  • Extracting text content from video
  • Speech to text transcription of video

Key services that are part of Azure Media Analytics include motion detection, face detection, emotion detection and optical character recognition.