Comments on the speaker overview paper

[wsstriving]

Dear all,
We are initiating a discussion on the paper "Overview of Speaker Modeling and Its Applications: From the Lens of Deep Speaker Representation Learning" here. Due to the page limitation (26 pages), there might be some topics that are not covered sufficiently in this paper.
We hope to draw inspiration from the valuable discussion, comments, and suggestions to improve the arXiv version (which has no strict page limitation). Especially, we are looking forward to receiving comments from industry professionals to uncover the real-world challenges and applications related to speaker modeling that are yet covered.
The comments can cover, but are not limited to, the following aspects:

1. Important topics that are not covered.
2. Important references that are missing.
3. The challenges that we didn't mention but are truly significant in real-world applications.
4. Possible questionable or outdated expressions.
5. Trends of this area in your opinion.
6. More.

Many thanks in advance!

[wsstriving]

update dataset info about voxblink2

[wsstriving]

update about the latest improvement of open-source toolkits

[Snowdar]

感谢分享SV领域的综述。只说知道的部分，提一个工具描述和表格部分的更正：asv-subtools 支持c++部署，同时kaldi只是特征提取和训练的可选工具，核心net部分一开始就是pytorch，也支持online dataloader使用python提取特征，因此也是容易做自定义的。

[wsstriving]

Will update the description the next version