Create the numerical labels – problem

[Norbert G]

hi Simon,

I have created full context labels and dumped time alignments there, created questions (180 only, commented out the suprasegmental question set in the script)
and now trying to create the numerical rep for the input to a DNN.
when running the script I got the following error msg:

2016-06-07 20:03:53,762 INFO labels : loaded /Users/s1574060/Documents/ss_dnn_letter_based/data/label_state_align/siedemwybranychopowiadan_8_orkan_00177.lab, 560 labels
2016-06-07 20:03:53,816 DEBUG labels : made label matrix of 1418 frames x 189 labels
2016-06-07 20:03:54,051 INFO labels : loaded /Users/s1574060/Documents/ss_dnn_letter_based/data/label_state_align/siedemwybranychopowiadan_8_orkan_00179.lab, 335 labels
2016-06-07 20:03:54,081 DEBUG labels : made label matrix of 832 frames x 189 labels
2016-06-07 20:03:54,229 INFO labels : loaded /Users/s1574060/Documents/ss_dnn_letter_based/data/label_state_align/siedemwybranychopowiadan_8_orkan_00180.lab, 315 labels
2016-06-07 20:03:54,258 DEBUG labels : made label matrix of 788 frames x 189 labels
2016-06-07 20:03:54,396 INFO labels : loaded /Users/s1574060/Documents/ss_dnn_letter_based/data/label_state_align/siedemwybranychopowiadan_8_orkan_00181.lab, 375 labels
2016-06-07 20:03:54,433 DEBUG labels : made label matrix of 1080 frames x 189 labels
2016-06-07 20:09:46,008 DEBUG acoustic_norm: MinMaxNormalisation created for feature dimension of 189
Traceback (most recent call last):
File “/Volumes/Network/courses/ss/dnn/dnn_tts/run_dnn.py”, line 957, in <module>
main_function(cfg)
File “/Volumes/Network/courses/ss/dnn/dnn_tts/run_dnn.py”, line 527, in main_function
min_max_normaliser.find_min_max_values(nn_label_file_list[0:cfg.train_file_number])
File “/Volumes/Network/courses/ss/dnn/dnn_tts/frontend/min_max_norm.py”, line 86, in find_min_max_values
features = io_funcs.load_binary_file(in_file_list[i], self.feature_dimension)
File “/Volumes/Network/courses/ss/dnn/dnn_tts/io_funcs/binary_io.py”, line 47, in load_binary_file
fid_lab = open(file_name, ‘rb’)
IOError: [Errno 2] No such file or directory: ‘/Users/s1574060/Documents/ss_dnn_letter_based/data/nn_no_silence_lab_189/siedemwybranychopowiadan_1a_orkan_00010.lab’

I am not sure if you can see it all, so I attached the copy in the file.

It looks like the script opens every single full context file in the label_state_align directory and creates its numeric representation and saves it to directory binary_label_189, which is fine. It iterates over all the files in that directory and then searches for another one, namely nn_no_silence_lab_189. I find it a bit confusing,since this directory is supposed to be created while doing the numerical labels, am I right?
In addition, when I went to search for it in Finder, the directory was actually there, so decided to run the script again but got the same error msg.
Could the reason behind it be that it is looking to create numerical values for suprasegmental features and can not find any lab files with them?

Attachments:

You must be logged in to view attached files.

[Srikanth R]

This is a bug which was corrected in latest version, please do svn update and you should be able to run without errors.

more specifically about the error:

1. In first step, the script creates its numeric representation and saves it to directory binary_label_189
2. In second step, it removes silence frames from directory “binary_label_189” and saves it to “nn_no_silence_lab_189”. But, if there are no silence segments in label file, it stops writing into this directory. Please comment the line numbers 69, 70 in “frontend/silence_remover.py” in order to correct the bug without updating the repository.

<pre class=”lang:python”>
#if len(nonsilence_indices) == frame_number:
# continue

[Norbert G]

hi Srikanth,

thanks for your reply which triggered another question, namely svn update.
I typed svn help in the command line which provides among many options update or up
argument, so I tried the below but get an error msg.

ppls-atlab-015:ss_dnn_letter_based s1574060$ svn update /Volumes/Network/courses/ss/dnn/dnn_tts/
svn: E155021: This client is too old to work with the working copy at
‘/Volumes/Network/courses/ss/dnn/dnn_tts’ (format 31).
You need to get a newer Subversion client. For more details, see
http://subversion.apache.org/faq.html#working-copy-format-change

Could you tell me what the correct syntax for the update is, please?

[Simon]

First, you cannot upgrade the copy in /Volumes/Network/courses/ss/dnn/dnn_tts/ – that is not owned by you!

You need to check out your own copy. If you still get the error, then do

$ svn upgrade

then

$ svn update

[Norbert G]

Hi,

Sorry, didn’t realise it, as I have never used svn before.
Decided to create my own copy then: svn co https://svn.ecdf.ed.ac.uk/repo/inf/dnn_tts.
When trying to run the file I get a msg that our old good friend bandmat is
not installed – will try to sort it now…
As for svn upgrade / update dont know what I am doing wrong but it doesn’t work for me.

[Author]

Posts