Updating x-vector readme to add diarization recipe link
#2
by
nauman
- opened
README.md
CHANGED
@@ -27,7 +27,7 @@ widget:
|
|
27 |
<iframe src="https://ghbtns.com/github-btn.html?user=speechbrain&repo=speechbrain&type=star&count=true&size=large&v=2" frameborder="0" scrolling="0" width="170" height="30" title="GitHub"></iframe>
|
28 |
<br/><br/>
|
29 |
|
30 |
-
# Speaker Verification with
|
31 |
|
32 |
This repository provides all the necessary tools to extract speaker embeddings with a pretrained TDNN model using SpeechBrain.
|
33 |
The system is trained on Voxceleb 1+ Voxceleb2 training data.
|
@@ -92,10 +92,19 @@ python train_speaker_embeddings.py hparams/train_x_vectors.yaml --data_folder=yo
|
|
92 |
|
93 |
You can find our training results (models, logs, etc) [here](https://drive.google.com/drive/folders/1RtCBJ3O8iOCkFrJItCKT9oL-Q1MNCwMH?usp=sharing).
|
94 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
95 |
### Limitations
|
96 |
The SpeechBrain team does not provide any warranty on the performance achieved by this model when used on other datasets.
|
97 |
|
98 |
-
#### Referencing
|
99 |
```@inproceedings{DBLP:conf/odyssey/SnyderGMSPK18,
|
100 |
author = {David Snyder and
|
101 |
Daniel Garcia{-}Romero and
|
|
|
27 |
<iframe src="https://ghbtns.com/github-btn.html?user=speechbrain&repo=speechbrain&type=star&count=true&size=large&v=2" frameborder="0" scrolling="0" width="170" height="30" title="GitHub"></iframe>
|
28 |
<br/><br/>
|
29 |
|
30 |
+
# Speaker Verification with x-vector embeddings on Voxceleb
|
31 |
|
32 |
This repository provides all the necessary tools to extract speaker embeddings with a pretrained TDNN model using SpeechBrain.
|
33 |
The system is trained on Voxceleb 1+ Voxceleb2 training data.
|
|
|
92 |
|
93 |
You can find our training results (models, logs, etc) [here](https://drive.google.com/drive/folders/1RtCBJ3O8iOCkFrJItCKT9oL-Q1MNCwMH?usp=sharing).
|
94 |
|
95 |
+
# Speaker Diarization with ECAPA-TDNN Embeddings
|
96 |
+
Note that, this trained x-vector model is also used for speaker diarization task. A full diarization pipeline including boudary preparation using RTTM files, speaker embedding extraction, and backend spectral clustering for [AMI dataset](https://groups.inf.ed.ac.uk/ami/corpus/) can be found [here](https://github.com/speechbrain/speechbrain/tree/develop/recipes/AMI/Diarization).
|
97 |
+
|
98 |
+
1. Run Inference for Diarization:
|
99 |
+
```
|
100 |
+
cd recipes/AMI/Diarization
|
101 |
+
python experiment.py hparams/xvectors.yaml
|
102 |
+
```
|
103 |
+
|
104 |
### Limitations
|
105 |
The SpeechBrain team does not provide any warranty on the performance achieved by this model when used on other datasets.
|
106 |
|
107 |
+
#### Referencing x-vectors
|
108 |
```@inproceedings{DBLP:conf/odyssey/SnyderGMSPK18,
|
109 |
author = {David Snyder and
|
110 |
Daniel Garcia{-}Romero and
|