Problem

[mstfsm]

Hello
I want to develop a Turkish TTS using GlowTTS but I have a problem during training. Can you help me for this problem?

! Run is removed from content/drive/MyDrive/Youtube/run-January-02-2023_07+06PM-f814d523
Traceback (most recent call last):
File "/usr/local/lib/python3.8/dist-packages/trainer/trainer.py", line 1591, in fit
self._fit()
File "/usr/local/lib/python3.8/dist-packages/trainer/trainer.py", line 1544, in _fit
self.train_epoch()
File "/usr/local/lib/python3.8/dist-packages/trainer/trainer.py", line 1308, in train_epoch
for cur_step, batch in enumerate(self.train_loader):
File "/usr/local/lib/python3.8/dist-packages/torch/utils/data/dataloader.py", line 628, in next
data = self._next_data()
File "/usr/local/lib/python3.8/dist-packages/torch/utils/data/dataloader.py", line 671, in _next_data
data = self._dataset_fetcher.fetch(index) # may raise StopIteration
File "/usr/local/lib/python3.8/dist-packages/torch/utils/data/_utils/fetch.py", line 61, in fetch
return self.collate_fn(data)
File "/content/TTS/TTS/tts/datasets/dataset.py", line 419, in collate_fn
mel = [self.ap.melspectrogram(w).astype("float32") for w in batch["wav"]]
File "/content/TTS/TTS/tts/datasets/dataset.py", line 419, in
mel = [self.ap.melspectrogram(w).astype("float32") for w in batch["wav"]]
File "/content/TTS/TTS/utils/audio/processor.py", line 477, in melspectrogram
D = self._stft(y)
File "/content/TTS/TTS/utils/audio/processor.py", line 528, in _stft
return librosa.stft(
File "/usr/local/lib/python3.8/dist-packages/librosa/core/spectrum.py", line 217, in stft
util.valid_audio(y)
File "/usr/local/lib/python3.8/dist-packages/librosa/util/utils.py", line 293, in valid_audio
raise ParameterError(
librosa.util.exceptions.ParameterError: Invalid shape for monophonic audio: ndim=2, shape=(119776, 2)

[albluc24]

from stack overflow it seems your files are in stereo. As far as I can see the first answer in there should work great, although it converts only a file at a time. You can either make a bash script to convert your files in batch, or else there are a miriad of ways to do that in general, so google is undoubtedly your friend here.