-
### Hi,
I want to add language ID embedding to my FastSpeech2 model.
The file says that i have to add this to my config file.
`tts_conf:
langs: 4 # Number of languages`
But I want to as…
-
-
Hello,
I am training a model in TalkNet. Before start training the spectrogram I created the durations.pt and f0s.pt
After that I started training the spectrogram with talknet_spect.py from exam…
-
I tried to finetune the vctk tacotron2+gst+xvector model on my dataset (5 speakers, ~1h each), but I met the following error:
```
Traceback (most recent call last):
File "/root/miniconda3/lib/p…
-
When I try to `install.py`, the build for `tacotron2` fails because it can't find `llvm-config`:
```
Error for /data/users/jamesreed/benchmark/torchbenchmark/models/tacotron2:
-------------------…
-
When installing `pip3 install neon_utils==0.15` getting error about dependency conflict
```
ERROR: numba 0.55.1 has requirement numpy=1.18, but you'll have numpy 1.22.2 which is incompatible.
ERRO…
-
We have to disable GPU train tests for a few models, because their original batch size doesn't fit into our current CI infra.
List of models that disabled train because of insufficient GPU memory:
…
-
Would it be possible to link one of the speech synthesis engines into the Linux speech dispatcher system? So that the default linux voice was neural network generated?
-
Not sure how to share this, but feel free to add.
https://drive.google.com/drive/folders/1jrMWhkK3KrOKsbPMYtCCcPczPmf8V-eQ?usp=sharing
It uses the default English encoder. To adapt tacotron text p…
-
## 🐛 Description
```
ValueError: No phonemizer found for language de.
raise ValueError(
File "/usr/local/lib/python3.9/site-packages/TTS/tts/utils/text/tokenizer.py", line 195, in init_fro…