Updates

Since our release this September, we have been collecting tons of valuable user feedback. Today, I am proud to announce a series of updates to Azure Media Indexer!

Features

• Recognizability
  • Now in addition to scores on the confidence of each word, we are also providing a general high-level “recognizability” score for each job so users can determine whether or not the provided media file was recognized by our language models. This can prevent the accidental inclusion of unusable output files from Indexer due to poor audio quality, etc.
• Java sample code
  • Special thanks to Andrew Weiss for contributing demo Java code here.

Fixes

• Unicode support
  • Users were having issues running Indexing jobs when input asset names or configuration xml files contained special characters. We have patched this limitation, so now all regular Unicode characters are supported!
• MIME type fixes
  • The SMI and TTML output files now correctly report MIME types as application/smi+xml and application/ttml+xml, respectively
• Minor Performance improvements

Feel free to continue sending feedback to us at indexer@microsoft.com! Also, stay tuned for updates with support for more languages. We are currently in a private preview for the Spanish language, with many more planned for the coming year.

Not sure what Indexer is?  Check out the introductory blog post here to learn more.

Indexer has been updated!  Check out the release notes for version 1.2 here!