I use chat.openai.com quite a bit for LilyPond. It almost never gives a
correct or directly useful answer, but often gives me ideas where I can
continue with LilyPond docs to figure out a solution. It's definitely
good at explaining how code fragments work if you paste them in. This is
a very similar situation to another esoteric language I use, OpenSCAD.
These languages don't benefit from anywhere near the amount of training
data that's available for languages like C++ and Python. That should
improve over time, but for now it's a crap-shoot.
Incidentally, I found Bing (supposedly based on GPT-4) is much worse
than chat.openai.com, despite its ability to look up info on the web,
and often just says it can't find anything.
Regards,
Curt
On 3/29/2023 3:43 PM, Saul Tobin wrote:
I've seen some examples of other people succeeding in getting ChatGPT
with GPT4 to compose simple music in other text based music formats.
I've had limited success getting it to output Lilypond code. It is
able to correctly structure the code with a score block, nested
contexts, and appropriately named variables, and bar checks at the end
of each measure. It seems to struggle to create rhythms that fit
within the time signature beyond extremely simple cases. It also seems
to struggle a lot to understand what octave pitches will be in when
using relative mode.
It also seems to have a lot of trouble keeping track of the
relationship between notes entered in different simultaneous
expressions. Just asking it to repeat back which notes appear in each
voice on each beat, GPT4 frequently gives stubbornly incorrect answers
about the music it generated. This makes it very difficult to improve
its output by giving feedback.
I'm curious whether anybody else has tried playing with this. I have
to imagine that GPT4 has the potential to produce higher quality
Lilypond output, given some of the other impressive things it can do.
Perhaps it needs to be provided with a large volume of musical
repertoire in Lilypond format.
