The region between performed notes was examined in nine instruments (flute, bass flute, piccolo, clarinet, oboe, bassoon, trumpet, violin, cello). On each instrument, eight intervals (2nd, 3rd, 5th, 7th, ascending and descending) and two playing styles (tongued/untongued, bow change/no bow change) were digitally recorded. Using a computer, the recordings were analyzed for time-varying power and time-varying spectrum. Analysis of as many as five recordings of a given interval and playing style showed that the performer could easily replicate a given transition, so the recordings were judged representative. In the tongued (bow change) case, the notes were farther apart, the amplitude dip between the notes (except in the cello), was greater, and spectral changes were more extensive than in the nontongued (no bow change) case. This pattern was not significantly influenced by variation in the size of the interval, the direction of the interval, or the instrument performing.
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.