I’m an industrial PhD student from the Swedish Agency for Accessible Media (MTM). Our mission is to provide people who have difficulties reading printed text, for example people with low vision or dyslexia, with text in other formats, such as Braille or talking books. We produce more than 100 Swedish newspapers and around 800 Swedish and English university textbooks yearly with text-to-speech synthesis (TTS).
My research interests lies in TTS for long and information-rich texts, and more specifically in the relation between the text in its written form and in its spoken form, for example:
- Input to neural TTS, e.g. levels of phonemic representations and pausing.
- Text normalization, pause and pronunciation generation as input to TTS.
I’m also interested in evaluation of long and information-rich texts read aloud.
My supervisors are Joakim Gustafson and Jonas Beskow.