The SpeechTek unit carries out research and develops technologies in the following application area:
- Speech recognition’ of audio streams (eventually coming from heterogeneous sources), "speaker verification/recognition" and "speech analytics" of e.g.: recording of telephone dialogues through “call centers”, Internet videos, analysis of both telephone or environmental tapping to furnish helping tools for investigations by law enforcement agencies, ecc.
- Assistive speech technology for language learning, to help language learning, measuring the levels of learning or the readings capabilities (especially for children, etc).
- Audio/video biometrics, with the objective of developing technologies for processing “personal” biometrics information, particularly focusing on wearable devices and/or egocentric data. Applications could be in the fields of security, and behaviour analysis/action recognition.