I transcribed the one hour audio, and it strongly suggests the honks aren't a delimiter since you end up with 9 to 18 bit stanzas. But not once do you see a 000 or 111. That's statistically very unlikely unless it's repeating a relatively short pattern. I've only got the bit stream transcribed, I still have to do the 5 bit sampling to see if it repeats. I'm less troubled by the missing 000 or 111 if it repeats.
Actually, stanza 8 is - - - _ _ . So yeah, you do have 000/111 (depending on if you think high is 0 or 1).
Also, chirp-set #7 (first one in stanza 3) is three high chirps in a row.
What I am noticing is that the chirp-sets in the first seven stanzas are only comprised of a limited set of words:
(a) _--
(b) --_
(c) -_- (annoyed gamer trying to solve this puzzle)
(d) --_-- (
very annoyed gamer trying to solve this puzzle)
(e) ---
Most, but not all, of these chirp-words are accompanied by either a high or low purr that happens just before, during, or after the word.
So for example in the first stanza we have the following sequence:
- high purr
- (a) + low purr
- (b) + high purr
- high purr
- (c)
- low purr
It seems pretty clear that the purrs somehow modify the chirp-words to expand their meaning somehow. If these are binary numbers, perhaps it's just another digit, or perhaps it's a negative or positive sign. If they are complex numbers, perhaps it's whether or not i is involved. Who knows.
I've mapped out all the chirps and purrs through #8, and I'm halfway through mapping the chords that accompany each section. It could be the chord changes indicate the beginning or ending of a given number.
But what is troubling to me is that the symbols from our main diagram do not all match the chirp-words. For example if the higher tone is equal to | in the diagram, then -|| matches (a), but, |-- has no match, -|- has no match, and --| has no match! Meanwhile if the lower tone is equal to | in the diagram, then |-- matches (a), -|- matches (c), --| matches (b), and only -|| has no match. Based on that it seems likely that at least when it comes to the chirps, the lower tone should be interpreted as | in the diagram. (That kind of makes sense, since the lower tone has longer wavelengths, and | is a longer symbol than -).
So then, to what can we match -|| (high-low-low)? If we use the purrs for that, then we'll have no problem, because the high-low-low pattern appears four times (if we ignore the barrier between stanzas) and three times if we don't. Even if we take purrs and nearby chirps together as words, we still have the high-low-low purr pattern occurring twice without interruption from chirps (if we ignore the barrier between stanzas) and once if we don't (at the beginning of stanza 5). This would make -|| the odd man out, but that might also explain why it has the two arcs symbol next to it: that could mean, "map this symbol to the purr words instead of the chirp words", if the double arc represents the purrs. And why not, since they are longer wavelength sounds?
The real oddities are stanzas 7 and 8.
Stanza 7 has only one chirp-word, (e), which means ZERO if this is binary. That could be an "end of message" marker.
Then stanza 8 has a weird chirp-word of _--- followed by ---- which are non-standard words according to the first seven.
Kloopy performed a similar analysis on a space recording of the UP. But I think he did not include the purrs or the chords into the equation, and he over-analyzed certain anomalies in the chirps, I think, misinterpreting as signal what was really noise in a few places (like his "L" symbols, which you can tell are a product of noise because the L bit only shows up on one of the two stereo channels).