yqzhishen
yqzhishen
The error message said ```text Missing input(s) for the inference session: tension, voicing ``` Please check your dsconfig.yaml in your dsvariance folder, or use the exporter-generated one to avoid misconfiguration.
Please use the right version of `pytorch-lightning` from the requirements.
Can the singer selection context menu be preserved to show recently used (or "pinned") voicebanks, and meanwhile add something like a `More...` button to open this dialog for detailed selections?
What model conversion project and what I am expected to do? We are non-profit team and may not be able to sponsor other projects, sorry
OK you can contact [email protected]
黄底叹号已经被 warning 给占用了，是框架提供给各插件用于提示用户的 API
gradio 是不是太旧了，升级一下 ```bash pip install -U gradio ```
目前需要使用DiffSinger数据集标注中的.ds文件进行训练，其他配置方法看代码和配置文件就行，训练和预处理代码还没妥善整理过
预处理binarize.py， 训练train.py，配置文件基本和diffsinger长得一样的，raw_data_dir是原始数据集，binary_data_dir是处理后数据集，数据集里面的.ds文件就note_seq, note_dur, note_slur三个key，放置方式和diffsinger也是一样的，其他的就结合代码和diffsinger的文档看吧
If you have ever made DiffSinger datasets you should be familiar with transcriptions.csv. See https://github.com/openvpi/MakeDiffSinger, if you haven't done that before and want to learn more details. There is also...