Exemplar-based speech waveform generation for text-to-speech: hybrid text-to-speech synthesis with Merlin

Here are audio samples for systems evaluated in the paper Exemplar-based speech waveform generation for text-to-speech:

@inproceedings{valentini18examplar,
  title     = {Exemplar-based speech waveform generation for text-to-speech},
  author    = {Cassia Valentini-Botinhao and Oliver Watts and Felipe Espic and Simon King},
  booktitle = {IEEE Workshop on Spoken Language Technology (SLT)},
  year      = {2018},
}

You can find code for replicating the proposed systems V-ES, T-EH, and T-ES here.