DSP Development Boards
So, our school came out with the new TI 64x DSP Testboards, pretty sweet stuff. Runs at 350 MIPS, those things are really quite marvelous. Finally something to work with.
I used the older boards last year in making a nonlinear adaptive speaker equalization system. With an accelerometer attached to a speaker's diaphragm, I used feedback from that accelerometer to determine how much the speaker's off from the original music/speech being played, and how I can make up for it by doing stuff before I pass them on to the speaker. Say if the bass is too quiet, then the system can turn the bass up a notch before passing it on to the speaker, so that it would come out equalized. This system needs to do this real-time, and that was a challenge. I had to use the newer TI 62x DSP boards then, implementing my nonlinear filtering algorithm was a pain because not many people have done it before. This one was particularly interesting, because my professor, Douglas Jones, had a student do a master's thesis on it with him, which is the filtering configuration I ended up using then.
I am looking to do some pattern classification tasks with these guys, such as pitch detection, timbre detection, so on and so forth. I don't know where the projects will go, but they might evolve into something.
I used the older boards last year in making a nonlinear adaptive speaker equalization system. With an accelerometer attached to a speaker's diaphragm, I used feedback from that accelerometer to determine how much the speaker's off from the original music/speech being played, and how I can make up for it by doing stuff before I pass them on to the speaker. Say if the bass is too quiet, then the system can turn the bass up a notch before passing it on to the speaker, so that it would come out equalized. This system needs to do this real-time, and that was a challenge. I had to use the newer TI 62x DSP boards then, implementing my nonlinear filtering algorithm was a pain because not many people have done it before. This one was particularly interesting, because my professor, Douglas Jones, had a student do a master's thesis on it with him, which is the filtering configuration I ended up using then.
I am looking to do some pattern classification tasks with these guys, such as pitch detection, timbre detection, so on and so forth. I don't know where the projects will go, but they might evolve into something.
