[FIXED] torchaudio "RuntimeError: Error loading audio file: failed to open file" for wav file

Issue

I was following this tutorial

However, when I ran the code under "Training", it gave me the following error.

RuntimeError: Error loading audio file: failed to open file /Users/leonardchoo/Desktop/dev_m1/audio_cnn/UrbanSound8K/fold10/30344-3-0-1.wav

The problem is, each time I run this, the error occurs at a different file.

error at...
# 1st run
fold2/106015-5-0-7.wav
# 2nd run
fold2/76086-4-0-22.wav
# 3rd run
fold9/14111-4-0-6.wav

I couldn't find anything like this on the web. I am genuinely confused by this.

My entire code can be found in this COLAB Notebook The dataset is from here

---------------------------------------------------------------------------
RuntimeError                              Traceback (most recent call last)
<ipython-input-14-2d83571e984b> in <module>
     59 
     60 num_epochs = 2   # Just for demo, adjust this higher.
---> 61 training(myModel, train_dl, num_epochs)

<ipython-input-14-2d83571e984b> in training(model, train_dl, num_epochs)
     19 
     20         # Repeat for each batch in the training set
---> 21         for i, data in enumerate(train_dl):
     22             # Get the input features and target labels, and put them on the GPU
     23             inputs, labels = data[0].to(device), data[1].to(device)

/usr/local/lib/python3.9/site-packages/torch/utils/data/dataloader.py in __next__(self)
    519             if self._sampler_iter is None:
    520                 self._reset()
--> 521             data = self._next_data()
    522             self._num_yielded += 1
    523             if self._dataset_kind == _DatasetKind.Iterable and \

/usr/local/lib/python3.9/site-packages/torch/utils/data/dataloader.py in _next_data(self)
    559     def _next_data(self):
    560         index = self._next_index()  # may raise StopIteration
--> 561         data = self._dataset_fetcher.fetch(index)  # may raise StopIteration
    562         if self._pin_memory:
    563             data = _utils.pin_memory.pin_memory(data)

/usr/local/lib/python3.9/site-packages/torch/utils/data/_utils/fetch.py in fetch(self, possibly_batched_index)
     42     def fetch(self, possibly_batched_index):
     43         if self.auto_collation:
---> 44             data = [self.dataset[idx] for idx in possibly_batched_index]
     45         else:
     46             data = self.dataset[possibly_batched_index]

/usr/local/lib/python3.9/site-packages/torch/utils/data/_utils/fetch.py in <listcomp>(.0)
     42     def fetch(self, possibly_batched_index):
     43         if self.auto_collation:
---> 44             data = [self.dataset[idx] for idx in possibly_batched_index]
     45         else:
     46             data = self.dataset[possibly_batched_index]

/usr/local/lib/python3.9/site-packages/torch/utils/data/dataset.py in __getitem__(self, idx)
    309 
    310     def __getitem__(self, idx):
--> 311         return self.dataset[self.indices[idx]]
    312 
    313     def __len__(self):

<ipython-input-8-8743d21efeae> in __getitem__(self, idx)
     32         class_id = self.df.loc[idx, 'classID']
     33 
---> 34         aud = AudioUtil.open(audio_file)
     35         # Some sounds have a higher sample rate, or fewer channels compared to the
     36         # majority. So make all sounds have the same number of channels and same

<ipython-input-7-f64618dd0374> in open(audio_file)
     13     @staticmethod
     14     def open(audio_file):
---> 15         sig, sr = torchaudio.load(audio_file)
     16         return (sig, sr)
     17 

/usr/local/lib/python3.9/site-packages/torchaudio/backend/sox_io_backend.py in load(filepath, frame_offset, num_frames, normalize, channels_first, format)
    150                 filepath, frame_offset, num_frames, normalize, channels_first, format)
    151         filepath = os.fspath(filepath)
--> 152     return torch.ops.torchaudio.sox_io_load_audio_file(
    153         filepath, frame_offset, num_frames, normalize, channels_first, format)
    154 

RuntimeError: Error loading audio file: failed to open file /Users/leonardchoo/Desktop/dev_m1/audio_cnn/UrbanSound8K/fold10/30344-3-0-1.wav

Solution

It was a silly trouble... i was missing /audio in the path...

I did not realize this was the case because the file it gave me error at always changed.

Answered By - Leonard