-
I'm looking to add 10-15 more speakers to the pretrained Libritts model of VITS (multispeaker). I have around ~20-30 minutes of each speaker data. I need help in implementing this on espnet2.
Any he…
-
Yi Liu, Hello.
Thank you very much for your solution!
I trained with dataset voxcelev1&2 and xvector_nnet_tdnn_amsoftmax_m0.20_linear_bn_1e-2_tdnn4_att. Everything works as expected. Training, …
-
My tests revealed that when using constant iterator tupled and zipped gives segfault.
Here is the code which shows that:
``` c++
#include
#include
#include
#include
#include
#include
#include
…
jpola updated
7 years ago
-
```
Warning messages:
1: Can't find iSEE in webR binary repository.
2: Can't find scater in webR binary repository.
3: Can't find scRNAseq in webR binary repository.
4: Can't find curl in webR bi…
-
Hi,
I've been using Kaldi in a way that probably isn't supported. I'm trying to train a dnn using a sequence of vectors as an input and a target of a label (0,1 or 2). Borrowing from the xvector trai…
-
from you code xvector-gpu.ipynb ,
the net construction is :+1:
Sequential(
(0): TDNN()
(1): TDNN()
(2): TDNN()
(3): TDNN()
(4): TDNN()
(5): StatsPooling()
(6): FullyConnect…
-
The `?scuttle::downsampleMatrix` documentation states:
> By default, the output format will be a dense matrix for an ordinary matrix ‘x’, and a sparse matrix for any sparse format (i.e., if ‘is_spa…
-
EDIT: 11.07.2024.
Discussion is about issue reported by [MohamedBelal999](https://github.com/MohamedBelal999) in the next comment.
Original issue:
There is quite a bit of small mistakes in http…
-
Hi,
First of all, very nice work.
well done.
I want to ask you about your xvector implementation.
I am not sure if I have a misunderstanding of how The Conv1D with dilation > 1 works.
Why did y…
-
Hello, thank you very much for sharing. The model has been trained. But As a newcomer in this field, I would like to ask you a question: assuming that I have trained a model, how can I use it to gener…