Rather than there being n buckets there should be log(n) buckets. Or rather each bucket should have a combination of buckets so that the lower frequency buckets have smaller frequency ranges within them. Example:
stream.get_audio_spectrum(buckets=50)
Will still return 50 buckets, however internally it will request more so that we have something like:
Out-Bucket 0 = In-Bucket 0
Out-Bucket 1 = In-Bucket 1 + 2
Out Bucket 2 = In-Bucket 3 + 4 + 5 +6
etc
This will mean that we get comparatively less bass response and much more high-range response, which I think is definitely necessary.
Rather than there being n buckets there should be log(n) buckets. Or rather each bucket should have a combination of buckets so that the lower frequency buckets have smaller frequency ranges within them. Example:
stream.get_audio_spectrum(buckets=50) Will still return 50 buckets, however internally it will request more so that we have something like: Out-Bucket 0 = In-Bucket 0 Out-Bucket 1 = In-Bucket 1 + 2 Out Bucket 2 = In-Bucket 3 + 4 + 5 +6 etc
This will mean that we get comparatively less bass response and much more high-range response, which I think is definitely necessary.