-
Hi all! I am now trying to configure some speaker recognition experiments using ECAPA-TDNN. However, when I was trying to load the model via yaml file, I got following error:
```
TypeError: ('Invali…
-
I am trying to run EER evaluations on a self-trained ECAPA model. I keep running in to the following error: "FileNotFoundError: [Errno 2] No such file or directory: '/network/datasets/restricted/voxce…
-
ResNet (e.g. https://github.com/alumae/torch-xvectors-wav)
ECAPA-TDNN (e.g. https://huggingface.co/speechbrain/spkrec-ecapa-voxceleb)
-
Thanks a lot for writing a script to prepare the AMI dataset - this is super useful!
One question: In most papers, results are reported on the AMI-IHM dataset, *e.g.*:
- https://www.researchgate.ne…
-
I think the paper of ECAPA-TDNN didn't mention any pre-processing on wav signal. And voxceleb_trainer didn't have this processing as well. Does it affect performance?
-
I ran **Speaker_Diarization_Inference.ipynb** notebook for getting diarization results on my audio data using the pretrained model. But as my data is quite noisy, the diarization results are not at as…
-
Hello,
I am training the ECAPA model from scratch using the provided [VoxCeleb recipe](https://github.com/speechbrain/speechbrain/blob/9df49e17c018c9617dfd3161a7db50a75086603e/recipes/VoxCeleb/Spea…
-
Hello,
When I try to run the urbansound classifier model with [instructions from the HuggingFace Hub ](https://huggingface.co/speechbrain/urbansound8k_ecapa) in Colab, I get the following error:
`…
-
hello lawlict, Thank you so much for the great work you shared.
My eer is 2.65% in voxceleb_test which can't reach to 1.*% when I training this model. I use aam-softmax loss which is same as loss men…
-
I find some errors in Voxceleb training experiments.
1. The valid error rate is 1 from the 11th epoch.
2.epoch: 1, lr: 7.46e-04 - train loss: 5.50 - valid loss: 1.56, valid ErrorRate: 4.64e-02
epoc…