Will Siri speak Frisian soon?

Language and speech technologist Henk van den Heuvel talks about speech recognition and atypical speech in the tenth episode of the podcast 'AI for Life'. The podcast series AI for Life is an initiative to offer a platform to experts from Radboud AI and other knowledge institutes in the Nijmegen region who work with data and algorithms.

Speech recognition has been around for a while and has made great strides. Think, for example, of Apple's Siri and Amazon's Alexa, which are used by millions of people every day. However, speech recognition still works best when spoken clearly or in a simple or specialised domain.

Dialect and stroke

The challenge is to take this even further so that speech recognition also works when the circumstances are more difficult. Think for example of spoken dialect, where the training possibilities of AI are often limited by smaller data sets. Or in cases of a speech impediment after a stroke. Another new direction is to detect emotions in speech and use them to better recognise the context of what is being said.

Atypical speech

Within the Radboud University and the Radboudumc, Telecats is working on more advanced forms of speech recognition in which atypical speech is also better recognised. In this episode of the podcast AI for life the following people were interviewed:
- Henk van den Heuvel (Centre for Language and Speech Technology)
- Sandra van Dulmen (Radboudumc)
- Arjan van Hessen(Telecats)

Platform

The podcast series AI for Life is an initiative to offer experts of Radboud AI and other knowledge institutions in the Nijmegen region who work with data and algorithms an accessible platform. Innovative entrepreneurs who convert this knowledge and expertise into useful products and services for society are also interviewed.

Listen to the podcast episode on Spotify (in Dutch)