geb@cadre.dsl.PITTSBURGH.EDU (Gordon E. Banks) (04/24/89)
We are interested in studying prosody (emotional tonality) of speech. We would like to analyze digitized utterances. I have the tools to digitize speech into sound files (at 11 or 22 kHz). I also believe I could write (or find) a program to Fourier analyze the waveforms to extract the dominant frequency vs. time (determinant of inflection). What I need is documentation on the way the digitized speech is stored in the sound file format. Can anyone give me the proper pointers to a TechNote or other reference? E-mail will be fine, since I doubt this is a subject of widespread net interest. Thanks.