I did a few tests today with the rhythm charts that I have constructed. One with the patterns from chart 1 where the note durations and onsets where derived from the number of 32nd notes (32 on the chart equals a whole note). Then I tried using the patterns in chart 2 where I used fractions of a whole note (32 equals one 32nd note). In general the second method sounds more natural. For example, the red curve on the chart makes a natural accelerando and ritardando, but this method also results in much more complex notation (alternatively, a higher degree of quantization).
There is one idea that I have for the piece that I want to be a musical version of a person telling a story or arguing for a case, and as he gets more excited, raising his voice and speaking faster and less and less articulated (that would be something like the Gaussian curve in chart 2). The musical version of this is a clearly stated motif that gets mingled up in slightly modified versions of the same motif, rhythmically displaced, and ends in a chaotic flurry of tones.
For this idea I will try to use versions of the patterns in chart 2 for the durations and onsets of the notes of the motif and use one of the curves in chart 1 for the rhythmic displacements of the modified motifs.