zeroQiaoba comments

Repositories
Issues
Comments

Results 5 comments of


                                            zeroQiaoba

nan value

@luoyetx @auroua @DL-Chang I also have this problem. How to solve this problem?

Can you open up your already trained models?

Since I am really busy recently, the model will be uploaded as soon as possible.

Can you open up your already trained models?

Our paper is avaliable in [link](https://arxiv.org/abs/1809.06225).

ivector pretrained model doesn't exist

Firstly, I check the link, it's avaliable, please check your web; Secondly, the 512-d xvector decribes the utterance-level speaker infomation, rather than frame-level infomation.

Varying the sound length

Yes. I have the same question. Maybe padding zero vectors in the end. But I do not know whether such a process will affect the performance.