nguyenlm comments

Results 23 comments of


                                            nguyenlm

is that able to train on Chinese dataset?

@jaywalnut310 This model is autoregressive or non autoregressive ?

is that able to train on Chinese dataset?

@jaywalnut310 Thanks you, I have some questions. 1. How about controllability ? 2. We can change the duration, energy or pitch ? 3. In the paper, you mentioned FastSpeech2 in...

Result getting worse when i use ground truth duration.

> @AlexanderXuan Thank you for your reply. I made some mistakes in my training, and when I fixed them and then train on my chinese dataset, the synthesized wavs are...

Result getting worse when i use ground truth duration.

@Liujingxiu23 Your config is same as the default ? What about training time to get 300K steps ?

Result getting worse when i use ground truth duration.

@Liujingxiu23, For me, with the sample rate 22050 it took about 8 days to get 180K steps

Result getting worse when i use ground truth duration.

@ductho9799 Hey bro, I've trained for Vietnamese.

Result getting worse when i use ground truth duration.

@ductho9799 Sorry, data is private so I cannot share it with you.

Result getting worse when i use ground truth duration.

@ductho9799 Hey bro, I think here is not the place for chatting, so please send me an email to [email protected], hope to here from you soon !

Result getting worse when i use ground truth duration.

@icyda17 Hi, For my exp, VITS is better than Fastspeech2 about prosody and quality. But in some cases VITS suffered from mis-pronunciation.

Result getting worse when i use ground truth duration.

> @leminhnguyen Thanks. Mis-puntutation in ur case means bad duration or tone issues? Btw, can I ask you more personally in private email or other chat platforms? You can contact...