Page 35 - AI Ready – Analysis Towards a Standardized Readiness Framework
P. 35
AI Ready – Analysis Towards a Standardized Readiness Framework
Open data available from [90], [91], and [92] are used along with private and generated data.
Multi-modal models with visual Question Answering combining images and questions to
predict answers and audio-visual speech recognition combining sound and video information
to identify speech content is used in this project.
The technology is deployed in smartphones so that people with disabilities can benefit from
the technology. This use case creates online and offline channels to collect feedback from the
users so that the performance of such a tool can be optimized.
28
