Random thoughts towards completion of my PhD thesis

So, now having finished 5.5 years of my PhD studentship, I am trying to consolidate all the ideas and work into a logical and presentable form for journal papers and thesis. I think any PhD thesis should tend to be a like a subspace, which itself is a vector space (closed under addition and scalar multiplication). It means that every element of the thesis and its combinations should be a part of the thesis.

Updated<08-01-16> So, the course registration for this semester is over and I had a sudden thought whether I should credit any new courses, initiated by the by a new upcoming rule in IISc, that a direct PhD student can get a M.Tech (Research) + PhD degree with a requirement of 21 credits as coursework. Even though I finished the credit requirement with 26 credits, it has been a long duration of 4.5 years since I credited a course. I thought it will be very good and exciting to credit courses. We get the feel of giving exams and assignments/ projects without the pressure of CGPA/ RTP as the  courses credited now will be counted as non-RTP additional courses. I think a research student should credit or be a TA for at least one course every semester so as to keep in touch with basics and learn new subjects helpful for our research. We may think that we can learn on our own through books/ video lectures but crediting and doing the course has a different higher level of commitment and understanding. So, even though I am in the sixth year, I am crediting two courses now.

Quality time for Research ?...?...

Its not a course work, an exam, a project or a contest. Its very different, research is like none of these activities. Research requires a self time, which I am now struggling to find. Distractions, mails, internet, talks and noises during the normal hours 10-5 p.m. In the 5th year of PhD, and need to schedule a quality time. Mornings and nights are only option. Nights are very lethargic and tiring mind for research, needs a fresh mind for research, mornings seem to be the only time. But, early morning is the only possible time. So, around 4 hours of quality time in the mornings is the best option, no disturbance in the dept or room. Maybe 6:30 am -7: 30 am, and then 8:30 am-11:30 a.m after breakfast, or 8-12 am . This time should be devoid of checking even dept. e-mail. The best part is the quality research for the PhD thesis is over early morning and have the rest of the day for other lab, dept. meetings talks, and extra-curricular activities. And night 8:30- 10:30 p.m. is also a optimum time after early dinner. 

Dull and boring part of Research..

As everyone thinks research is very interesting, innovative and tough, it is sad that sometimes we tend to do something which is already done with slight changes, complexify it and publish and get publications. And as we say it, we tend to do it ourselves. Like taking an already attacked application, already used methods do some permutation of different methods and think we are doing great research. May it will be much better if we try to think afresh about a new, unattacked  application and devise a simple method to solve it, even if it is very simple. Given that we do lot of literature survey, we get many new ideas to use them, but we should not get carried away by what others have already done. We may get inspired from others work, but for our own research we must have our own different, novel ideas. As it is always not possible that our new ideas work, we tend to do this dull and boring part of research. Hope we will get over it and make it more interesting.

Effect of high frequencies in audio signals and intricacies in audio research

A nice article describes why high sampling rates are not required for audio. The audio signal is the most intriguing of all signals. It is quasi-periodic but non-stationary over long  periods above 20 ms.

I have been evaluating instants of maximum excitations or epochs and I was using 3 Core i5 systems simultaneously as I had to the run the same code using different parameters to get the best performance and accuracy. I have re-refined my output recursively so as to get the best accuracy and identification rates.

On the random way to research

My another blog describes my early stages of my research. I am well into my 3rd year of my PhD, having redefined, refined, re-refined my problem statement. Having gone through a lot of research papers, journals related to my problem, some very much theoretical, application oriented and some seemly useless papers.

The problem regarding my problem statement is the solution is a random entity. The problem is seemingly NP-hard. The real problem is it is difficult and challenging. The problem statement is: Given a single channel audio signal containing mixture of only two sources: 1) speech, 2) non-speech, the output should be two separated channels one containing only speech and other non-speech signal. Now, its a supervised learning where dictionaries for both speech and non-speech need to be learnt before testing on a mixed audio.

Now, I have seen lot of good papers on sparse dictionary learning, having finally formulated the problem which is quite difficult and has lot of applications if I solve or solve to some extent. Most of the past dictionary learning has been used for object tracking in videos, image classification, few for speech recognition  and denoising but very few for source separation.

Having done the literature survey, where most of the good papers are published   in IEEE Conference on Computer Vision and Pattern Recognition (CVPR), IEEE Transactions on Signal Processing,  International Conference on Machine Learning, IEEE Workshop on Machine Learning for Signal Processing and Journal of Machine Learning Research. 

I see source separation of guitar music and other sources as one aspect of my problem. A recording of a guitar tune played by me: