SAIL LABS Releases Media Mining Indexer 4.2

VIENNA, Austria - SAIL LABS Technology released Media Mining Indexer 4.2, the latest version of its product for broadcast new indexing.

First released in 2002, the Media Mining Indexer enables indexing of live news feeds and multimedia archives, producing annotated text output to be used in digital asset management systems for subsequent multimedia search and retrieval. The SAIL LABS Media Mining Indexer is built on SAIL LABS' proprietary speech processing and language technology:

  • Automatic speech recognition for conversion of speech-to-text
  • Speaker Change Detection to segregates audio segments based on speaker for focused information retrieval
  • Speaker identification classifies who has been saying what in the news.
  • Topic classification helps locating stories relating to user specific interests.
  • Named entity detection makes finding relevant information easier by highlighting words within the text output that belong to certain categories (e.g. persons, locations, organizations).

The Media Mining Indexer 4.2 extends the product's existing features with:

  • Improved latency of the speech recognition
  • Major speed improvement in Named Entity Detection
  • Availability of an integrated package using Microsoft Installer for Indexer, Language Model Tool, Feeder, API (this is an alternative to the traditional zip)
  • Internet License Activation allowing customers to activate their licenses irrespective of SAIL LABS office hours
  • Speed and footprint improvement in Language Model Tool generated models
  • Enhanced language model handling (up to 15 percent, depending on the language)
  • Improved documentation and structure of the API and header files
  • Fixes for rare problem conditions
  • Improved logging and problem determination capability (both in terms of quality and quantity)

The Media Mining Indexer 4.2 is available as a free update to licensed customers with corresponding Support & Maintenance Contracts from SAIL LABS and its global network of partners.


SpeechTek Covers
for qualified subscribers
Subscribe Now Current Issue Past Issues