egrinstein / neural_srp Goto Github PK

Hi, I've tried to reproduce neural-srp and encountered some problems:

1. when I proprocess the tau-nigens dataset, the following error occurs, and when I change the value of self._nb_unique_classes from 2 to
   3 it have been solved, Does it mean that the samples in the dataset do not have more than 2 sound source activities at most?

tnb_classes[frame_ind, active_event] = 1
IndexError: index 2 is out of bounds for axis 1 with size 2

2. But When I tried to visualize the tau using the neural-srp-multi.bin, it reported that:

target_doas = target_doas.view( target_doas.shape[0], target_doas.shape[1], 3, max_nb_doas ).transpose(-1, -2)
RuntimeError: shape '[1, 50, 3, 2]' is invalid for input of size 500

Is it because I change the value of self._nb_unique_classes?

3. When I tried to loading the doanet.bin to visualize the tau, it occured:
   
   how can I solve it?

Thank you in advance, looking forward to your reply!

Hello, after reading your code, i have some questions:

In GccExtractor, the feat matric shape is

gcc_feat = np.zeros((nb_frames, self._nb_bins, n_output_channels))
gcc_feat = gcc_feat.transpose((0, 2, 1))

and, in Preprocessor::extract_all_feature, the feat is trans to

        feat = self._gcc_extractor(audio_in)
        nb_frames = feat.shape[1]
        feat = feat.transpose((0, 2, 1)).reshape((nb_frames, -1))

which not make sense. ( i mean, the nb_frames refers to different meanings

I can't get it, could u help me about it ？ If there is something wrong in this version of code? Thanks!!!