Speech:Spring 2014 Mitchell Dezak Log


 * Home
 * Semesters
 * Spring 2014
 * Proposal
 * Report
 * Information - General Project Information
 * Experiments - List of speech experiments

Week Ending February 4th, 2014
Review past class logs
 * Task:

2/1
 * Results:

Looked at past logs to see what other groups did in the first week. Mostly quick, simple stuff.

Get a basic understanding of what our group is supposed to accomplish and get on board with everyone. Nothing so far.
 * Plan:
 * Concerns:

Week Ending February 11, 2014

 * Task:


 * Results:


 * Plan:


 * Concerns:

Week Ending February 18, 2014
To find the genTrans scripts and make sure that they are up to date and could use any improvements. I will also read about them on the genTrans page on the wiki. I want to know more about the data backup and speech corpus.
 * Task:

2/16
 * Results:

Read up on the speech corpus and genTrans. In logs from last year it was said that there might be some missing data from some of the scripts. It looks as though, the missing data was added to that script later on. Perl was also used to do this, which I'm not too familiar with, so depending on what I am able to find I might have to add some missing data using Perl as well. Looking at what was done, it doesn't look too complicated but it is still something that I am not familiar with.

2/17

Looking to download the genTrans file to my computer to review it. I know this was done last year and some issues were run into, so hopefully those were resolved.

Access and make any updates that are necessary to the genTrans. It looks like getting familiar with Perl might have to happen sooner rather than later.
 * Plan:


 * Concerns:

Week Ending February 25, 2014
2/20- Read some logs and wrote a couple tasks for the week below.

2/23- Read up about Perl, starting to understand it better. Analyzed some of the code the best that I could.

2/24- Kept analyzing the genTrans, thinking of changes to be made.

2/25- Logged in, read logs


 * Task:
 * Get a basic understanding of Perl
 * Once I get a basic understanding of Perl, start to work out the genTrans

2/23- So far I am noticing that Perl isn't any different from other programming languages. I can't say that I am that surprised. That doesn't mean that I am able to make out all of the genTrans, but it is progress. Luckily to make things easier, the genTrans does have explanations of the code throughout. Some of it is a bit vague, so parts of it is still not entirely easy to understand.
 * Results:

for example:

$message =~ s/sw[0-9]*[A-B]-ms98-a-[0-9]* [0-9]*.[0-9]* [0-9]*.[0-9]* //; # remove everything before the message

That is from genTrans6.pl The code seems pretty confusing at first glance, and the comment doesn't tell me very much. There is more code like this to look into.

I have also noticed that in some of the genTrans that some of the code has been commented out. A lot of it has been in 5 and 6 actually.

from genTrans5:

#$message =~ s/\/.*?]//g; #$message =~ s/\[noise\]//g; #remove [noise] #$message =~ s/\[laughter\]//g; #remove [laughter] #$message =~ s/\[vocalized-noise\]//g; #remove [vocalized-noise] #$message =~ s/\[laughter-//g; #remove [laughter- #$message =~ s/\[.*?\///g; #remove [ / $message =~ s/(\si\-\s)/ i /g; #replace i- with i     $message =~ s/<.*?>//g; # remove < > #$message =~ s/-\[//g; #remove -[ #$message =~ s/\]-//g; #remove ]- #$message =~ s/\]//g; #remove ] #$message =~ s/\[//g; #remove [

There is probably a reason behind it, (maybe something that I have forgot about), but I thought it was something worth noting.

2/24- After noticing that the genTrans5 and 6 files had commented out a lot of stuff, I went back and looked at genTrans4. It has the same messages without them being commented out. I found out on the page: http://openitware.org/projects/index.php/Speech:GenTrans that it is done because the messages that are commented out cause data to be lost. Therefore, that script did not work.

I want to be able to understand Perl so that I can understand what I am looking at on the genTrans, then I can also work out everything in those files. I don't want to jump in too quickly without knowing what I am doing otherwise I can end up wasting a lot of time. Trial and error doesn't hurt, but I'd like to reduce error as much as possible. Nothing at the moment
 * Plan:
 * Concerns:

Week Ending March 4, 2014
2/27- Logged in, read logs

3/2- Discussed with Jared and John over Email what we need to work on as a group and separated them individually.

3/3- Worked on Poster Proposal

3/4- redid poster proposal


 * Task:

I have the Poster Proposal to work on, I'm not entirely sure what I'm going to do for this, but I plan on getting it started and we can finalize it as a group.


 * Results:

3/3- I worked on the Poster Proposal in Power Point, because that seems to be the easiest way to put the contents of what you want to put on a poster on a computer. Finding graphics to use for this seems to be the hardest part, since we are essentially working off of a command line most of the time.

3/4- I scrapped what I had before and wrote the abstract for the Poster proposal.


 * Plan:


 * Concerns:

I'm not entirely sure what the Poster Proposal is about

Week Ending March 18, 2014
3/15- logged in


 * Task:


 * Results:


 * Plan:


 * Concerns:

Week Ending March 25, 2014
3/20- Logged in. Read up on running experiments.

3/23- wrote in tasks and plans for the remaining week.


 * Task:

Learn about running experiments before running experiments.

Run my first experiment this week.


 * Results:

3/23- Read through the model building steps to get a run down again of running an experiment. David walked me through running one during Wednesday's class. I have also gone through some logs that have experiments that were done. I'm going to start my first experiment tomorrow.

3/25- I tried running experiments 0238 and 0239. 0238 had a senome of 1000 and came up with the error: BEGIN failed--compilation aborted at RunAll.pl line 48. I think tried experiment 0239 with a senome of 3000 and came up with the same error. Not exactly sure how to fix this. I went through the wiki pages again and haven't found anything that says I should be doing anything different so I must be missing something.


 * Plan:

3/23- I will attempt to run my first experiment tomorrow. Results to follow.


 * Concerns:

Week Ending April 1, 2014

 * Task:


 * Results:


 * Plan:


 * Concerns:

Week Ending April 8, 2014

 * Task:


 * Results:


 * Plan:


 * Concerns:

Week Ending April 15, 2014

 * Task:


 * Results:


 * Plan:


 * Concerns:

Week Ending April 22, 2014

 * Task:


 * Results:


 * Plan:


 * Concerns:

Week Ending April 29, 2014

 * Task:


 * Results:


 * Plan:


 * Concerns:

Week Ending May 6, 2014

 * Task:


 * Results:


 * Plan:


 * Concerns: