Inhoud blog
  		• Friday, 14/12/12
  		• Friday, 7/12/12
  		• Tuesday 4/11
  		• Tuesday 27/11
    Zoeken in blog

    Beoordeel dit blog
      Zeer goed
      Goed
      Voldoende
      Nog wat bijwerken
      Nog veel werk aan
     
    Keyboard
    keyboard
    dit is een blog voor peno3
    14-12-2012
    Klik hier om een link te hebben waarmee u dit artikel later terug kunt lezen.Friday, 14/12/12
    Today we finished our report.

    14-12-2012 om 16:35 geschreven door keyboard123
    07-12-2012
    Klik hier om een link te hebben waarmee u dit artikel later terug kunt lezen.Friday, 7/12/12
    We worked on our paper, so it is as good as done. The only thing we are missing is a graph. we've also been working on our program so we would have a presentable version for our demo-day. We've been experimenting with more noise reduction and implemented a cross-correlation algorithm to our program.

    07-12-2012 om 17:49 geschreven door keyboard123
    04-12-2012
    Klik hier om een link te hebben waarmee u dit artikel later terug kunt lezen.Tuesday 4/11
    Today we worked on our paper, but it is not finished yet. We like to get it finished the next session so we could send it to our promotors. 
    We made our poster for the demo session. We reorganised our methods again to test it for the demo-day (We made a new method to find our hitpeak). We found out out that Cepstrum worked good (against all odds). But our demonstration is not as good as we tought, we work hard to get it better.

    04-12-2012 om 18:00 geschreven door keyboard123
    30-11-2012
    Klik hier om een link te hebben waarmee u dit artikel later terug kunt lezen.
    We started working today an hour early, at 13:00.
    The most of our time we spent in improving and expanding our paper. We processed some of our data so we could visualise it in our report.
    We have also worked on a different approach to record our samples. Instead of using a recorder on our computer, we will record everything using playrec itself. This could improve our results quite a bit more.

    30-11-2012 om 17:00 geschreven door keyboard123
    27-11-2012
    Klik hier om een link te hebben waarmee u dit artikel later terug kunt lezen.Tuesday 27/11
    Today we worked hard on our paper, we worked on text, figures, cleaning algorithms for the appendices, etc.
    We also gathered succes rates from all our different functions. First we did this for all our keys and then with only the five and four best keys. 
    From these figures we saw that the top three function are (from three to one) : fft_single_PCC -> fft_PCC_VGL -> fft_single_freq_PCC
    These functions we will use in playrec.
    We are also still working on a better function for finding the hitpeak. We already tried a lot of things but no function seems to be working properly.

    27-11-2012 om 18:05 geschreven door keyboard123
    23-11-2012
    Klik hier om een link te hebben waarmee u dit artikel later terug kunt lezen.Friday 23/11/2012
    Today, we got feedback on our brochure and corrected it directly (thus completing our last version). We optimised several algorithms and tested it with playrec. The results were good but they can be better. That is it! 

    23-11-2012 om 17:49 geschreven door keyboard123
    20-11-2012
    Klik hier om een link te hebben waarmee u dit artikel later terug kunt lezen.Tuesday 20/11/2012
    We worked on the brochure (finished it) and the report. Someone worked out a function which doesnt use the average anymore, but it calculates the algebraic distance to each training key and the letter with the smallest distances, should be correct. This method isnt that succesfull, a rate about 40% with some very good exceptions. 
    Someone kept optimizing the sound recognition function. 
    We also started a new approach for calculating key patterns using the standarddeviation and the average. With these specifications we make an interval which can be used as a template.
    writing to optimize the search of the hitpeak, because our earlier function sometimes used the releasepeak (when the releasepeak is higher than the hitpeak).

    20-11-2012 om 17:55 geschreven door keyboard123
    09-11-2012
    Klik hier om een link te hebben waarmee u dit artikel later terug kunt lezen.Friday 9/11/2012
    We had to refresh our database, because of the new microphone. Luckily we didnt have to split the samples in to seperate keystrokes, because the function to find keys is close to find them all. This made our work much easier (and mostly quicker).
    We have used our final program, that combines all the other (find key,find peak, recognize which key), in playrec, and it worked!
    Also we kept improving our report, using more statistics and figures,...
    And lastly we researched cepstrum, so we might be able to implement it into our program and test it properly.

    09-11-2012 om 17:37 geschreven door keyboard123
    06-11-2012
    Klik hier om een link te hebben waarmee u dit artikel later terug kunt lezen.Tuesday 6/11

    We made an algorithm that uses the quefrency domain for finding a correlation using the matlabcode rceps. Working with rceps we found out that the correlation between the signal and the averages from each letter is raised but it isn’t able to find the right letter.

    We combined two algorithm: the peakfinder and PCC.

    We worked on the report. 

    06-11-2012 om 18:00 geschreven door keyboard123
    23-10-2012
    Klik hier om een link te hebben waarmee u dit artikel later terug kunt lezen.Tuesday 23/10
    We worked on our paper. We created the structure of it and worked out some of its topics. We also rearranged our functions, and added some comments to make them more readable. Finally, we continued to work on the algorithm that seperates the keys from the sound.

    23-10-2012 om 18:03 geschreven door keyboard123


    Archief per week
  		• 10/12-16/12 2012
  		• 03/12-09/12 2012
  		• 26/11-02/12 2012
  		• 19/11-25/11 2012
  		• 05/11-11/11 2012
  		• 22/10-28/10 2012
  		• 15/10-21/10 2012
  		• 08/10-14/10 2012
  		• 01/10-07/10 2012
    E-mail mij

    Druk op onderstaande knop om mij te e-mailen.
    Gastenboek

    Druk op onderstaande knop om een berichtje achter te laten in mijn gastenboek


    Blog als favoriet !


    Blog tegen de wet? Klik hier.
    Gratis blog op https://www.bloggen.be - Meer blogs