Commit
•
f3d9c66
1
Parent(s):
5824c21
Update README.md
Browse files
README.md
CHANGED
|
@@ -26,6 +26,10 @@ widget:
|
|
| 26 |
|
| 27 |
This repository provides a pretrained ECAPA-TDNN model using SpeechBrain. The system can be used to extract speaker embeddings as well. It is trained on Voxceleb 2 development data only.
|
| 28 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 29 |
# Pipeline description
|
| 30 |
|
| 31 |
This system is composed of an ECAPA-TDNN model. It is a combination of convolutional and residual blocks. The embeddings are extracted using attentive statistical pooling. The system is trained with Additive Margin Softmax Loss.
|
|
|
|
| 26 |
|
| 27 |
This repository provides a pretrained ECAPA-TDNN model using SpeechBrain. The system can be used to extract speaker embeddings as well. It is trained on Voxceleb 2 development data only.
|
| 28 |
|
| 29 |
+
| Release | EER(%) | minDCF(0.01) |
|
| 30 |
+
|:-------------:|:--------------:|:--------------:|
|
| 31 |
+
| 05-03-21 | 1.45 | 0.17 |
|
| 32 |
+
|
| 33 |
# Pipeline description
|
| 34 |
|
| 35 |
This system is composed of an ECAPA-TDNN model. It is a combination of convolutional and residual blocks. The embeddings are extracted using attentive statistical pooling. The system is trained with Additive Margin Softmax Loss.
|