The main goal of the SPINDLE project is to generate automatically a set of keywords using natural language processing techniques from the automatic speech-to-text transcriptions of the university podcasts to increase the open educational audio and video resources discoverability. These automatic transcriptions are generated using state-of-the-art Large Vocabulary Continuous Speech Recognition Software (CMU Sphinx and the Speech Analysys tool from Adobe Premiere Pro). Please check our blog for more information and updates.
Mining a Year of Speech
CMU Conquest Spoken Dialogue System
Dihana Spoken Dialogue System