SpeechSplit
SpeechSplit copied to clipboard
auspicious3000
Unsupervised Speech Decomposition Via Triple Information Bottleneck
Greetings, thanks for such a good project. In my experiment, i used the same dataset VCTK as yours, and i had only trained for 68000-steps. The log of my experiment...
Hello!It’s not long for me since I just learned voice conversion.So I have many questions, and one of which is how to appoint the source speech and the target speech...
After I tested a demo in my own data, I found that the content of the generated speech was not converted, why is this? Looking at the comparison chart, the...
I have two questions which maybe are just of low-level. What do "R"、"F"、"U" mean in demo.py respectively？ And how can I obtain the content embeding，pitch embedding， rhythm embedding?
Hi. Thank you for the fantastic project. Does your model is capable to transfer content, rhythm, and pitch between different sentences? I've prepared a demo.pkl file in the way that...
The loss of my training set looks normal, but the loss of the validation set has been rising. The loss of my training set looks normal, but the loss of...
Hello, thank you so much for the code and paper! I'm trying to train the model on [speech command data](http://storage.googleapis.com/download.tensorflow.org/data/mini_speech_commands.zip). I've made the train and validation data sets through 2...
I don't understand how to get alignment when the input(utterance) to the rhythm-encoder is different from inputs(utterance) to pitch/content-encoders. ps(I don't understand the implementation details of variant in Appendix B.3)....
If the length of content code, rhythm code and pitch code is different from each other, how do they align since there is no attention mechanism in decoder?
Metadata
Owner
Metadata
Unsupervised Speech Decomposition Via Triple Information Bottleneck