Fig. 4From: A unit selection text-to-speech-and-singing synthesis framework from neutral speech: proof of conceptCorpus (a) and target (b) vowel duration and F0 distributionsBack to article page