Synthesising short vowels from their long counterparts in a concatenative based text-to-speech system.