Fix typo in melscale_fbank (#3487)

Summary:
Resolves https://github.com/pytorch/audio/issues/3486

Pull Request resolved: https://github.com/pytorch/audio/pull/3487

Differential Revision: D47724733

Pulled By: mthrok

fbshipit-source-id: 26f5641a8271a7e50c4a33861d09b0c8274b29e4

torchaudio/functional/functional.py

@@ -549,7 +549,7 @@ def melscale_fbanks(
         meaning number of frequencies to highlight/apply to x the number of filterbanks.
         Each column is a filterbank so that assuming there is a matrix A of
         size (..., ``n_freqs``), the applied result would be
-        ``A * melscale_fbanks(A.size(-1), ...)``.
+        ``A @ melscale_fbanks(A.size(-1), ...)``.
 
     """