speechbrain VS imgaug

You likely have a very broken sample in one of your batches. It looks like your training actually went through a few batches before it horked the error at you. A quick google shows a similar issue in the github repo: https://github.com/speechbrain/speechbrain/issues/649 .

https://github.com/ggerganov/whisper.cpp https://speechbrain.github.io/

I don't know if it's the best, but Speechbrain is supposed to be state of the art.

- https://github.com/speechbrain/speechbrain

There are many good options when it comes to tools and libraries for implementing data augmentation into our deep learning pipeline. You could for instance do your own augmentations using NumPy or Pillow. Some of the most popular dedicated libraries for image augmentation include Albumentations, imgaug, and Augmentor. Both TensorFlow and PyTorch even come with their own packages dedicated to image augmentation.

https://github.com/aleju/imgaug This one is way better for image.

I have to improve my dataset with more backgrounds featuring obstacles. At the moment I'm working on creating a dataset with both items and trinkets, and I'm planning on using https://github.com/aleju/imgaug which will replace most of the stuff I'm doing with PIL.

Do you acknowledge that there is ImageDataGenerator and ImgAug?

Albumentations no longer uses the imgaug library by default. All previous imgaug augmentations in the library are reimplemented in Albumentations with the same API (but you can still install Albumentations with imgaug if you need the old augmentations).

I would also recommend you to give a try to TensorFlow Object Detection Model - https://github.com/tensorflow/models/tree/master/research/object_detection with augmentation - https://github.com/aleju/imgaug pipeline. The same worked for me in a similar use case where I had to localise logo on documents.