-
I want to train ljspeech.tacotron2.v3, gpu 1.
But after 119 (from 200) epochs wav file sounds like this - https://www.sendspace.com/file/egdzmj.
Is it normal?
Thanks in advance.
-
seem my fastspeech2 implementation can't handle long sentence in some dataset such as KSS. FOr Ljspeech and other dataset from other person report that it's still fine. I'm thinking about the maximum …
-
I unfortunately do not have the computing resources to train a MB MelGAN model with VCTK.
My other option is to use the one from https://github.com/kan-bayashi/ParallelWaveGAN, and it is working ni…
-
Hello,
I am new to espnet and TTS in general, so sorry if this question might not make too much sense. If I got some basic things wrong, don't hesitate to point it out please.
For a privat project…
-
when I train parallel wavegan with parallel_wavegan.v1.debug.yaml, it works fine. (training audio is 16khz so I had to fix some hyperparameters and I'm using yesno repo)
however, when i change yaml i…
-
Hi, I'm trying to install `espnet` with the following command
```
make KALDI=~/kaldi/ PYTHON=~/.pyenv/versions/3.6.8/envs/espnet_env/bin/python3.6 CUPY_VERSION='' -j 4
```
And came across this e…
-
Hi, as Eren requested, this is an issue to follow progress of the training a larger PWGAN model for multiple speakers.
-
Dear Team,
I am using transliterated text for my custom language/dataset in EspNet TTS.
The used transliteration is phonetic, but not G2P canonic. Therefore I kept
`trans_type="char"`
in run.s…
-
I used mandarin demo (transformer tts & parallel-wavegan) in
`https://colab.research.google.com/github/espnet/notebook/blob/master/tts_realtime_demo.ipynb#scrollTo=MoNYASQ-A0cN`
I set device to c…
-
Hi,
just found https://arxiv.org/pdf/2005.05106.pdf
It seems to provide significantly better quality than regular MelGAN, and is also stunningly fast (0.03 RTF on CPU). The authors will be publi…