-
hi,
is it possible to train a non-English data?
-
Hope to receive your reply.
-
@Rayhane-mamah, I'm currently starting work on using a WaveNet based Vocoder to connect to Tacotron-2. Any pointers to your Tacotron-2 implementation to start with? I have a wavenet_vocoder implementa…
-
In your [https://r9y9.github.io/wavenet_vocoder/](https://r9y9.github.io/wavenet_vocoder/)。
you said you trained Tacotron2 (mel-spectrogram prediction part): trained 189k steps on LJSpeech dataset (P…
-
@r9y9 Thank you for sharing PR!
which mode to generate spectrograms file in Tacotron2? As "--conditional=" in Wavenet_vocoder.
![1111111](https://user-images.githubusercontent.com/42974031/4547…
-
> Exiting due to Exception: Incompatible shapes: [2,12799,10] vs. [2,12799]
[[Node: model/loss/sub_7 = Sub[T=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:GPU:0"](model/loss/strided_sli…
-
Hi there , im training a Polish speech model , spectrograms looks quite good, even the sound in /Log category is not so bad like for 35k steps. Even thoo I can't synthesize any audio , from text file …
-
Hey, I just noticed a likely bug in line 376 in `wavenet_vocoder.feeder._adjust_time_resolution` where it says `x=audio.trim(x)` rather than `x=audio.trim_silence(x, self._hparams)`.
-
I'm new to taco and I'm really interested because I want to make a voice 'mask' for a specific TTS (Loquendo in Spanish), and I need to know if the inference speed is rasonable
I don't need the con…
-
When I run the code just for synthesis using the model I downloaded from the link provided, I get the following error:
Traceback (most recent call last):
File "synthesis.py", line 204, in
…