From 3af7f976a207153cc12c9752dbcce1dbb9024520 Mon Sep 17 00:00:00 2001
From: moto <855818+mthrok@users.noreply.github.com>
Date: Tue, 6 Oct 2020 21:40:40 -0400
Subject: [PATCH] [doc] Update librosa link

---
 torchaudio/transforms.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/torchaudio/transforms.py b/torchaudio/transforms.py
index ef0a58e637..2e7e4f37ee 100644
--- a/torchaudio/transforms.py
+++ b/torchaudio/transforms.py
@@ -198,7 +198,8 @@ def __init__(self, stype: str = 'power', top_db: Optional[float] = None) -> None
 
     def forward(self, x: Tensor) -> Tensor:
         r"""Numerically stable implementation from Librosa.
-        https://librosa.github.io/librosa/_modules/librosa/core/spectrum.html
+
+        https://librosa.org/doc/latest/generated/librosa.amplitude_to_db.html
 
         Args:
             x (Tensor): Input tensor before being converted to decibel scale.