After a long puase, I have now resumed working on my programming project to implement and test a speaker independent speach recognition algorithm suggested by Huang and Kuh, 1992. Having only allowed for soundfile playback in the early versions, I have now added support for real time audio input. I decided to make use of the PortAudio I/O library. After fiddling around with it for a few hours, I must say it really works well and is extremely easy to work with as compared to Apple’s native API which I find utterly confusing for anything other than the built in audio hardware.

Click the tag/category for related posts