Prajwal

No, it was reported on the entire test data. What @Rudrabha means is that, you can listen to a few generated samples and see if they sound alright. Usually, it...

Hi, you would need a different preprocessing script, where you specify in each frame which part of the image to save as a "crop". In our evaluation script, for example,...

> ValueError: Multi-valued hyperparameters cannot be empty: all_images Ensure that this line is returning the list of all face images. If not, your path is incorrect.

The number of hours is calculated based on (total number of images / fps) / 3600. Ensure all the face images are detected by the script.

Does your test set only contain a single video? Please check if any exceptions are being silently ignored in the code

there is a minimum frame limit = hparams.T. Please add print statements in different places and see if your videos are being skipped. > C:\Users\Graham\Desktop\Lip2Wav-master\synthesizer\saved_models\logs-01\taco_pretrained\tacotron_model.ckpt-5000.data-00000-of-00001 yes.

Yes, please @ authors, it would really help us to train new models and not be bottlenecked by the expensive k-means step.

In file dataloader.py line 241, change num_workers to a small enough number, and after that, I did not run out of memory.

That's surprising. Could you please uncomment the following lines in scripts/sudokuExtractor.py: > Line 19, 21 23 and run it again to see the outputs after `preprocessing`, after `cropping out sudoku...

Hi @ymittal, The neural network used to classify digits is not good enough for real life images. I had implemented this project with a vanilla neural network so that I...