Executedone/Chinese-FastSpeech2
基于标贝数据继续训练,同时对原本的FastSpeech2模型做了改进,引入了韵律表征以及韵律预测模块,使中文发音更生动且富有节奏
Python
Issues
- 1
The bert details
#24 opened by mondorysix - 1
- 1
支持中英文混合训练或推理吗
#23 opened by aidway - 1
RuntimeError: The size of tensor a (33) must match the size of tensor b (36) at non-singleton dimension 1
#17 opened by ltydd - 0
有合成的样例吗
#19 opened by lucasjinreal - 0
您好 我正在写论文, 我想知道您是怎么样在原来的模型上进行微调的?
#18 opened by ZaneLeo111 - 3
- 0
TextGrid和expanded_embeds.pkl
#16 opened by ltydd - 0
size mismatch
#15 opened by Bingtai1015 - 2
多speaker韵律预测?
#8 opened by zjwang21 - 2
FileNotFoundError: [Errno 2] No such file or directory: './preprocessor/expanded_embeds.pkl'
#7 opened by xiaoru168 - 2
- 0
- 2
- 1
About fine-tuning issues.
#11 opened by ltydd - 0
- 8
- 2
如何支持流式tts输出
#6 opened by zhaolibo1989 - 1
有和成例子吗?
#1 opened by lucasjinreal - 1
使用的标贝数据的标注是数据自带的还是MFA提取的呢?
#5 opened by MaxMax2016 - 1
No module named 'transformers'
#2 opened by wizardk