Speech recognition technologies.
Broadcasting, media monitoring, call centers, health care and reporting
The PerVoice Speech-to-Text Engineprovides speaker-independent transcriptions in multiple languages. The system can use very large vocabularies and is robust enough to recognize and ignore many kinds of background noise, making it useful in non-studio conditions, as well. PerVoice transcription solutions add value to traditional audio or Audio/Visual files by permanently synchronizing text with audio. It is possible to create Audio/Visual indexes and search for keywords, easily correct and certify transcriptions, automate subtitling and closed-captioning processes, find out what customers are thinking when they talk to a Call Center operator, and analyze content for word use and frequency. You can also integrate PerVoice products with other upstream and downstream IT solutions.