buriburisuri / speech-to-text-wavenet

Speech-to-Text-WaveNet : End-to-end sentence level English speech recognition based on DeepMind's WaveNet and tensorflow

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

cant find the assets/data directory???

iamsidharthbnair opened this issue · comments

Sir,can you help me finding the assets/data directory to put the corpus..... after installation of all dependencies........

after you downloaded the VCTK corpus - just create such a directory or make a symlink to the VCTK path there

@uralik Thank you.

@sidh104
IOError: File asset/data/speaker-info.txt does not exist -> means that file is not there, may be you should double check that.

as I see you created another dir VCTK inside train folder. this is wrong. Just check the code.

you should make dir data inside asset, and data should already have the content of VCTK-Corpus files.

we did and now we get this error @uralik @buriburisuri

root@ubuntu:/home/varsha/Desktop/project/speech-to-text-wavenet-master# python train.py
Traceback (most recent call last):
File "train.py", line 26, in
data = VCTK(batch_size=batch_size)
File "/home/varsha/Desktop/project/speech-to-text-wavenet-master/data.py", line 45, in init
labels, wave_files = self._load_corpus(data_path)
File "/home/varsha/Desktop/project/speech-to-text-wavenet-master/data.py", line 111, in _load_corpus
self.max_len = np.max([len(s) for s in sents])
File "/usr/local/lib/python2.7/dist-packages/numpy/core/fromnumeric.py", line 2252, in amax
out=out, **kwargs)
File "/usr/local/lib/python2.7/dist-packages/numpy/core/_methods.py", line 26, in _amax
return umr_maximum(a, axis, None, out, keepdims)
ValueError: zero-size array to reduction operation maximum which has no identity
root@ubuntu:/home/varsha/Desktop/project/speech-to-text-wavenet-master#

@sidh104 you need to extract VCTK-Corpus.tar.gz directly in /asset/data . Are you using pre-trained model?

@fazalWahid56 we r using the VCTK-corpus and v dnt have any pre-trained model other than that

commented

@sidh104 if I use the pre-trained model and I got the error : IOError: File asset/data/speaker-info.txt does not exist .
what should I do?

@chrislinan u shud put the pre-trained model in asset/data directory just put all the files and folders which is inside VCTK-corpus into assets/data directory