-
It's in the line 178. Why does it do like this?
-
In the speaker_verification_plda.py, I do iteration in the set_loader. It will auto-c…
-
Hello everyone,
I am currently trying to build a model for speaker verification using my own data.
To do so, I duplicated the VoxCeleb recipe and created my own CSV files.
### Training the x-ve…
-
As the HDL-generation part has had a big overhaul, it would be great if also the Xilinx-FPGA path can be tested.
As I mainly use Intel (Altera) FPGA's, I do not have the ability for the moment to che…
-
when I do the speaker embedding cluster（step 4）, the job was failed and I look up the fail log. It shows that “OSError: ./data/magicdata160h_dev_test/dia_part/embedding/CTS-CN-F2F-2019-11-15-****.seg …
-
![Snipaste_2021-08-21_00-12-01](https://user-images.githubuserco…
-
I try to train the standard xvector model on VoxCeleb1 trainset using the script `runVoxceleb.sh` with 4 GPUs. And I completely use the default parameters in `runStandardXvector-voxceleb1.py` except f…
matln updated
3 years ago
-
The documentation does not specify the purpose of the "VoskRecognizer.**SetWords**(bool)" method.
What does it do?
There is also no example of using the SpkModel object. Is there a valid code exa…
-
I used this toolkit on some long audio files. The clustering stage takes long time, like a couple of hours for a 1.5 hour audio (full of speech). Is this normal? Is it possible to run the clustering a…
-
The format is not hdf5 or npz using kaldi so I am not sure what to change in the code.
Thank you very much!