-
hello I have a problem extracting x-vector using :
`EncoderClassifier.from_hparams(
source="speechbrain/spkrec-xvect-voxceleb",
savedir="pretrained_models/spkrec-xvect-voxceleb",
)`
w…
-
Is there a way to match an encoding to the voxelceleb database. Analyze a voice and know which celebrity it is? Thanks
All I could see was comparing two voices with encodings.
Currently:
```
…
-
Hello~~
I'm trying to run PLDA. I found that when I set the value of latent_topics bigger than 1 (like 5), and then call .make_doc() function. It posts "Segmentation fault (core dumped)" bug.
When …
-
Stage a.1: Generating pseudo-speakers for libri_dev_enrolls.
Computing PLDA affinity scores of each source speaker to each pool speaker.
cut: exp/models/2_xvect_extr/exp/xvector_nnet_1a/anon/xvector…
-
Hi,
Today I tried to use the lidbox and I ran into the following error:
`
File "D:\anaconda\envs\slid\lib\site-packages\lidbox\embed\sklearn_utils.py", line 6, in
from plda import Classifier …
-
Hi,
I've used your speaker_id module to train a model on a custom dataset. Initially the `n_classes` parameter has been set to 4 in `train.yaml`. Now I would like to increase this parameter and add…
-
Hi dears
I want to use Kaldi pre-train speaker verification models.
suppose I have two arrays or something else that contain ivectors of two voice (e.g. a=[1,1,1] b=[2,2,2]) and I want to compute pl…
-
I have successfully trained speaker Identification model using my custom dataset. However, I want to ask is there any way in your toolkit that ifI want to add a new speaker to my enrollment without do…
-
I'm getting this error after running `fast_PLDA_scoring`.
```
scores_plda = fast_PLDA_scoring(enroll_xvectors_stat, test_xvectors_stat, ndx, plda.mean, plda.F, plda.Sigma)
print(scores_plda.score…
-
You only need to use one of those PLDA models for your system. Also, if you have enough in-domain training data, you'll have better results training a new PLDA model. If your data is wideband micropho…