-
前几天下载了sherpa-onnx-1.8.8, 运行里面的范例 vits-zh-aishell3
```
python3 ./python-api-examples/offline-tts.py
--vits-model=./vits-aishell3.onnx \
--vits-lexicon=./lexicon.txt \
--vits-tokens=./tokens.t…
-
跑样例过程中,在Synthesizing这一步中报错,请问是哪里的问题
multiple speaker fastspeech2!
Traceback (most recent call last):
File "/home/ubuntu/yqk/PaddleSpeech/paddlespeech/t2s/exps/fastspeech2/../synthesize.py", line …
-
使用master分支代码,在aishell3数据集上训练,使用单卡时,损失正常;使用2卡时,反向传播一次后,loss为nan
-
我用aishell1+aishell3重新训练了lamma,top5准确率在90%以上。但在推理时,合成音频的内容与text完全不同,且内容是没有意义的
-
### Describe the bug
After 130000 steps of vits multi-speaker training with aishell3, the generated wav belongs to one speaker voice regardless of the speaker_idx. I have debugged the inference proce…
-
### bug描述 Describe the Bug
采用多进程时,paddle抛出异常,但是程序可以正常运行,这个异常虽然不会影响程序结果,但是为什么会出现这样的情况呢?
最小复现代码:
```python
import multiprocessing
from tqdm import tqdm
import paddle
class custom_dataset(pa…
-
https://modelscope.cn/datasets/speech_tts/AISHELL-3/summary
这里提到的数据集格式中的 interval 文件如何得到的?
-
(torch12cu113) PS E:\project\MockingBird> python .\pre.py E:\project\MockingBird
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or tr…
-
hi,
have you taken chinese pinyin tones into consider when training the model .
for example aishell3, ['chi2','qi3','hong2','ying1','qiang1'], were you process the data only in the way ['chi','qi',…
-
im sorry for mi incomptence but im a newby, im trying but im unable to make the apk run, i followed all steps but im not able to get out of the UnsatisfiedLinkError. what am i doing wrong? i have the …