Vecsys Research develops speech processing technologies for
multilingual, large vocabulary speech recognition (speech-to-text),
automatic audio segmentation, language identification and speaker
recognition. Core speech recognizer engines are available for
broadcast data and conversational speech in multiple languages.
This core technology can serve as the basis for a variety of
applications ranging from interactive conversational systems to the
automatic indexing of audio data.
For the latter class of applications, large vocabulary continuous
speech recognition is the key technology for enabling content-based
information access in audio and video documents. Most of the
linguistic information is encoded in the audio channel of audiovisual
data, which once transcribed can be accessed using text-based tools.
Among the most common applications of our technology are audio
and audiovisual data mining (broadcast data, call center data), media
monitoring, media asset management, and telephone-based conversational
systems.
