January 5, 2005

Stereo beginnings

The hardest part of this project for me is going to be the electronics interface. I've taken one EE course in digital electronics, and I've toyed around with some simple circuits before using do-it-yourself type books, but my theoretical knowledge is close to zero, so I'm not very good at coming up with circuit designs on my own. I do have the ability to learn new things, though, so I'm hoping I can pick up whatever I need.

Continue reading "Stereo beginnings"

Posted by mill1991 at 12:09 PM · Stereo control

January 4, 2005

Early progress in speech recognition

Since my area of research here at the U is natural language processing (NLP), that is the part of the project I am best equipped for, and thus expect the most early progress on. In fact, I have already made some preliminary progress. I'm going to be using the Sphinx system developed at Carnegie Mellon. It's biggest drawback as far as I can tell is that it uses an HMM trigram language model as opposed to a structural language model. However, since the speech commands I intend to use are probably not even grammatical sentences (e.g. "Window up", "Volume down"), and are generally fairly short (the longest is probably something like "Rear left window down"), a trigram model probably captures most of the information needed.

Continue reading "Early progress in speech recognition"

Posted by mill1991 at 1:44 PM · Speech recognition

January 3, 2005

Project Description

The goal of this project is to "hack" my car to include a speech recognition system. I think my first goal should be to be able to control my stereo via voice control. If this should work out, my next task would be to control more encapsulated functions like the heating/cooling system and windows. The purpose of this blog is to track my progress on this project. I don't expect to update extremely frequently, as school will be starting soon and I have other research. Depending on how fast the early stuff can be done, I may have to delay the bulk of this project until summer, due to actual research and classes (I know, they're such a drag). But insofar as I do make progress, I will update this site with it. So, if you're actually interested, keep an RSS feed (howto) and don't bother checking until I update.

The tagline of this site is "Tracking the failure of my latest pie in the sky project." Sometimes I start projects that are a little too ambitious for my limited time, finances, and intelligence. One reason for keeping a project log (plog) is that with my potential failure or success out in the public, I may be more motivated to actually complete the project. The worst case scenario is that I'll have a record of my project, and I might be able to tell where I went wrong or right.

Posted by mill1991 at 12:44 PM · General information