Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[WIP] generate egs for age, emotion or TS task #2211

Open
wants to merge 1 commit into
base: master
Choose a base branch
from

Conversation

LvHang
Copy link
Contributor

@LvHang LvHang commented Feb 10, 2018

@pegahgh
I cleaned up my repository. Now the codes and scripts focus on your age, emotion or teacher-student task. Bear in mind, these codes make the egs in the same archive have distinct lengths and use "ranges in script-file lines" method. Please check the codes and see whether you have any other requirements.

In the comment of #2169 , you mentioned "The idea is that in future, specially the label can be even vector like posteriors (e.g. TS learning application)" for function "ProcessUttToLabelFile()".
According to my knowledge, the loss function of Teacher-Student Training is
loss = - sum_all_state {teacher_output(state, t) * log student_output(state, t)}
That means, for each frame, we need a posterior vector.
I think to supply this requirement, we simply modify the "ProcessUttToLabelFile()" is not feasible. Function "ProcessUttToLabelFile()" only uses utt_id to find the label so that we can use the label generates the one-hot vector later. However, to slove the problem, we need provide another file which contains the output of teacher system in matrix format or posterior format. If possible, please show me a file which you can provide and I will figure out a way to slove it.

Best,
Hang

@stale
Copy link

stale bot commented Jun 19, 2020

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

@stale stale bot added the stale Stale bot on the loose label Jun 19, 2020
@stale
Copy link

stale bot commented Jul 19, 2020

This issue has been automatically closed by a bot strictly because of inactivity. This does not mean that we think that this issue is not important! If you believe it has been closed hastily, add a comment to the issue and mention @kkm000, and I'll gladly reopen it.

@stale stale bot closed this Jul 19, 2020
@kkm000 kkm000 reopened this Jul 19, 2020
@stale stale bot removed the stale Stale bot on the loose label Jul 19, 2020
@stale
Copy link

stale bot commented Sep 17, 2020

This issue has been automatically marked as stale by a bot solely because it has not had recent activity. Please add any comment (simply 'ping' is enough) to prevent the issue from being closed for 60 more days if you believe it should be kept open.

@stale stale bot added the stale Stale bot on the loose label Sep 17, 2020
@danpovey
Copy link
Contributor

@LvHang I'm not sure if you want to finish this? It seems to lack an example script.

@stale stale bot removed the stale Stale bot on the loose label Sep 17, 2020
@LvHang
Copy link
Contributor Author

LvHang commented Sep 18, 2020

@danpovey Sure, I will revisit the code and pegah's paper about age/TS. Then, I will provide an example script.

@stale
Copy link

stale bot commented Nov 17, 2020

This issue has been automatically marked as stale by a bot solely because it has not had recent activity. Please add any comment (simply 'ping' is enough) to prevent the issue from being closed for 60 more days if you believe it should be kept open.

@stale stale bot added the stale Stale bot on the loose label Nov 17, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
stale Stale bot on the loose
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

3 participants