Error message wording when n_fft is larger than length of the input

BEFORE POSTING A BUG REPORT Please look through existing issues (both open and closed) to see if it’s already been reported or fixed!

Describe the bug The error/warning message when n_fft is larger than the length of the input seems to be the wrong way around. Currently, I get n_fft=2048 is too small for input signal of length=512. Shouldn’t it be n_fft=2048 is too *large* for input signal of length=512?

To Reproduce

import librosa
y = librosa.tone(220, length=512)
librosa.stft(y, center=False)

Expected behavior A clear and concise description of what you expected to happen.

Screenshots If applicable, add screenshots to help explain your problem.

Software versions* INSTALLED VERSIONS

python: 3.8.3 (default, Jul 2 2020, 16:21:59) [GCC 7.3.0]

librosa: 0.8.0

audioread: 2.1.8 numpy: 1.19.1 scipy: 1.5.2 sklearn: 0.23.1 joblib: 0.16.0 decorator: 4.4.2 soundfile: 0.10.3 resampy: 0.2.2 numba: 0.50.1

numpydoc: None sphinx: None sphinx_rtd_theme: None sphinxcontrib.versioning: None sphinx-gallery: None pytest: None pytest-mpl: None pytest-cov: None matplotlib: 3.3.0 presets: None

[bshall]

@miltonG-isla, @bneslyn21 This is issue is just about the wording of the error message. The actual code (including the > sign) is doing the correct thing and shouldn’t be changed.

If you want to get rid of the error you should either use a smaller value for n_ftt or pad your input.

[bmcfee]

Ah good catch. I think this is purely cosmetic (the behavior is correct, but the description is wrong). But we should fix it for the next release.