Machine-learning system tackles speech, object recognition at once

26 September 2018 | United States
MIT computer scientists have developed a system that learns to identify objects within an image, based on a spoken description of the image. Given an image and an audio caption, the model will highlight in real-time the relevant regions of the image being described.
You are not logged in. For existing users, please login to learn more about this premium content. For new users, please register with idtechex.com or use the links below for more details about premium content from IDTechEx.

Subscription Benefits


A subscription from IDTechEx allows you to access our full body of research content on emerging technologies, including market, technology and company data, alongside direct engagement with our expert analysts.
Subscription content includes:
  • Market research reports
  • Forecasts
  • Premium articles
  • Company profiles
  • Industry digests
  • Webinars
  • Innovation maps
  • Event summaries and write-ups