Apache Tika is a toolkit for detecting and extracting metadata and structured text content from various documents using existing parser libraries. You can find the latest release on the download page. See the Getting Started guide for instructions on how to start using Tika.