Rafael Valle

Have you trimmed silences from beginning and end of the audio files?

You can stop training once you confirm that the validation loss is clearly going up. In this model above, is the attention good around 400k right before you decrease the...

Please try starting with the LJSpeech model and training on LJS and your data. Copy the value of the LJ speaker embeddings to the 0-th embedding of the new speaker...

setting the attention prior to True will help the model learn attention must faster and allows for training at least 2 steps of flow at the same time. https://github.com/NVIDIA/flowtron/blob/master/config.json#L34

Yes, we're looking into it!

An alternative way to change speed is to train Flowtron with a token duration predictor, modifying the attention mechanism to work the hard alignments and then scale the durations during...

We'll soon make Flowtron Parallel available https://twitter.com/RafaelValleArt/status/1281268833504751616?s=20

does it repeat itself in the middle of the sentence or just at the start? if it's just at the start then it's a gating issue, possibly the gate loss...

you can fine tune the model with `use_cumm_attention` and checking if you get better results. another option is to always use the attention prior, i.e. during training and inference. other...

Where's the tensorflow dependency ?