Now that people are asking their smartphones about everything from billiards to baking, perhaps it's time we took a peek at speech recognition software.
Indeed, the technology has been around for decades, so speech applications have gotten pretty sophisticated -- which is great fun for users, but can create challenges for developers who are still working with older, flat text files. This week, author Colin Beckingham describes how you can bridge that gap by moving toward open source standards in speech processing. Learn how to use Python to convert existing flat lexicon files to the Pronunciation Lexicon Specification (PLS) XML format, and then convert PLS files back to flat text format. This can increase your cross-project compatibility, robustness, and extensibility in areas like Unicode.
It can also enhance your appreciation of your digital assistant the next time you ask her for a traffic update.
Until next week,
John Swanson and the developerWorks editorial team
Our other top features on developerWorks this week:
- Getting grips with fpm: Tightening SUID programs (AIX and UNIX)
- Explore what customers should expect in cloud service standards (Cloud computing)
- Use the RTVDSKINF command to avoid disk storage disasters on IBM i (IBM i)
- Query social media and structured data with InfoSphere BigInsights (Information Management)
- Explore Linux as a scientific computing platform: Crunch numbers with Sage and Python (Linux)
- Configure DOORS and Rational Team Concert for globally distributed workers (Rational)
- Complement canvas with HTML markup: Blend the canvas API and HTML/CSS model (Web development)
- New WebSphere Developer Technical Journal (WebSphere)