Hi, thanks for this impressive work. I am reproducing your work on the SVS task, I saw that the phoneme information is included in that. Can I have a question where can we get that phoneme, is that from the DNN-HMM model as you mentioned in the paper for the TTS task?
Hi, thanks for this impressive work. I am reproducing your work on the SVS task, I saw that the phoneme information is included in that. Can I have a question where can we get that phoneme, is that from the DNN-HMM model as you mentioned in the paper for the TTS task?