Roeland Ordelman and Arjan van Hessen
Netherlands Institute for Sound and Vision and University of Twente, The Netherlands

Objectives

In spite of significant efforts and progress, automatic speech recognition (ASR) is not yet a tool that scholars can easily deploy in their research.
In our project we focus explicitly on those aspects that are crucial for scholars with respect to actually deploying the technology:

• Usability of the technology: working with an ASR engine and using its output for research
• Sustainability: the longer-term availability of the technology for scholars, with state-of-the-art performance, maintained, updated, and accessible.