-
Notifications
You must be signed in to change notification settings - Fork 62
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Getting started with a custom dataset #34
Comments
Hi, sorry for the late reply. |
Hi, thank you for your response. Is it ok to use the SCV2_Dataset or DESED_Dataset? I see they are regular dataset classes, which are better for loading my standard audio WAV files. What is the difference from SEDDataset? Can HTS-AT support multi-channel audio wav? Thank you. |
Hi, SCV2 is for speech command v2 dataset, and desed is for the sound even detection dataset, and ESC is for the ESC-50 dataset. I think SED dataset for SCV2 might be the best fit from which you can change it into your own dataset. Yes, it is possible to support multi-channel audio, but first you might need to change the first layer to map more than one channel to the deep feature--> meaning that the pretrained model is no longer workable. Another way is that you can merge multi-channel into the single-channel, or performing the classification on multi-channel and take their average results. |
Hi,
Thank you. |
Hi,
Thank you for your excellent work!
I want to use HTS-Audio-Transformer for my custom dataset, different classification task.
Are there any instructions on how to run the model for a different dataset? From which file should I start?
Thanks
The text was updated successfully, but these errors were encountered: