Information Content of a Sound Spectrogram
The measurement of the information content of a sound spectrogram is approached on the basis of the number of distinguished marking regions in the sample space provided by the spectrogram. After first considering the information content of a general Cartesian diagram, the upper limit of information content in a piece of sound spectrogram is derived. It is shown that there can be an exchange of information content between the frequency and time coordinates, depending on the relationship of the analyzing filter bandwidth and the base period of the signal processed. A general existence theorem is stated for displaying a time- and frequency-limited signal on a section of sound spectrogram. Finally, it is shown that the average information content of a -voice print,- as these are currently produced, can be expected to be somewhat less than that of the usual sound spectrogram.
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member and would like to subscribe to the E-Library then Join the AES!
This paper costs $33 for non-members and is temporarily free for AES members.