http://googleresearch.blogspot.ch/2016/02/text-to-speech-for-low-resource.html

This is the third episode in the series of posts reporting on the work we are doing to build text-to-speech (TTS) systems for low resource languages. In the first episode, we described the crowdsourced acoustic data collection effort for Project Unison. In the second episode, we described how we built parametric voices based on that data. In this episode, we look at how we are compiling a pronunciation lexicon for a TTS system.