tacotron
tacotron copied to clipboard
keithito
A TensorFlow implementation of Google's Tacotron speech synthesis with pre-trained model (unofficial)
Do you have suggestions to resume training from a checkpoint? How can i modify code to do that? @keithito
I want to deploy this model in the tensorflow serving, but saved-model is needed. It seems complex and difficult for me to figure the raw inputs and final outputs of...
I want to use this code for a low resource language(pashto) so guide where I should start and what to do I am beginner in programming. Thanks in advance
how to shift this project to tensorflow 2.0 how to convert tf.contrib.training.HParams in 2.0? and how to export this model for tensorflow serving
I have a 2 GB GTX 750 Ti, when I try to train on it I get an OOM error. Evidently the memory is not enough. Since I cannot afford...
I am trying to modify the tacotron model for urdu language. I was modifying the symbols.py file and since urdu is a cursive language, if I remove spaces from between...
I am trying to train the model on the Urdu Data set which is of 2 hours on 400K Steps . I have trained with `cleaners=transliteration_cleaners` I do have the...
Currently this project is working on tensorflow 1.3.0. For synthesizing the voice we are trying to run this on tensorflow-gpu as mentioned in [this](https://github.com/keithito/tacotron/issues/15#issuecomment-329054549) . But after installing tensorflow-gpu==1.3.0,while we...
Hi, I am currently using one GPU 1080 Ti, if I put in another GPU card would it benefit? Wanted to get an opinion before I made the investment. thanks...
https://github.com/keithito/tacotron/blob/d26c763342518d4e432e9c4036a1aff3b4fdaa1e/text/numbers.py#L57 You need to replace dash character to space, too. `_inflect.number_to_words(num, andword='', zero='oh', group=2).replace(', ', ' ').replace('-', ' ') ` Without that, 91 => ninety-one.
