Technology overview

Veridict’s speech technology provides the next step in speech recognition for state-of-the-art speech-enabled systems and applications.

The award-winning speech technology from Veridict is optimized for time-critical online applications, where recognition output must be generated with minimum latency. The system operates and decodes the speech while the user speaks and applies advanced predictive methods to enable near-instant output. Although users can use any speaking style or rate, and speak rapidly and continuously without making any pauses between words, the system still detects and outputs the individual words incrementally with a fraction of a second delay. This enables the system to start reacting to user input events before the full utterance has ended.

For additional information, please see the links to the left.