search

athena-team / athena

an open-source implementation of sequence-to-sequence based speech processing engine
https://athena-team.readthedocs.io
Apache License 2.0
954 stars 197 forks source link

Can the authors provide pretrained weights?? #340

Closed paulista5 closed 3 years ago

paulista5 commented 3 years ago

Hi, we trained our own version of speech simclr, using streaming convnets(Facebook) as an encoder. We wanted to see the similarity of the sentences spoken by different speakers. In the first case, we took the same sentences spoken by different speakers and measured the similarity. Similarly, in the 2nd case, different sentences were spoken by different speakers. We also measured the similarity between different sentences spoken by the same speaker. However, in our testing, we found that the similarity measures were all over the place for our model. We want to do a similar test on your model, using pre trained weights.

Can the authors provide the pre-trained weights? Thanks

stale[bot] commented 3 years ago

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

stale[bot] commented 3 years ago

This issue is closed. You can also re-open it if needed.