Changjin Han

[FastPitch] Why do you hierarchically predict the variance features (pitch and energy)?

2

Thank you always for sharing your thoughtful code. As we can see in FastPitch code, you added the pitch embedding to encoder output before passing the energy predictor. https://github.com/NVIDIA/DeepLearningExamples/blob/da7e1a701bd44885c5537afa7974be391f82401e/PyTorch/SpeechSynthesis/FastPitch/fastpitch/model.py#L300 Why...

Where to add an attention prior (betabinom prior)

3

Hello! I have a question about the adding position of an attention prior. You added the attention prior before calculating forwardsum loss like this. https://github.com/imdanboy/jets/blob/44e3dbcb9e7e5368158917748fa2c6b45039b4d0/espnet2/gan_tts/jets/loss.py#L147 It can decrease the forwardsum...

Total duration of training dataset

1

Thank you for your excellent work! Could you share the duration of the language-specific datasets and the overall total duration used to train OpenVoice V2? For instance: - English: 10K...

Total duration of training dataset

2

@jasonppy Thanks for sharing the open-source and enhanced version. But I am also curious about the amount of dataset. Could you let me know how many hours of data was...