Datasets:
Updated Dataset Card for clean subsets
Browse files
README.md
CHANGED
|
@@ -38,7 +38,7 @@ where $T$ is the length of the audio, $P\left(Y^{\text{aligned}} \mid X\right)$
|
|
| 38 |
|
| 39 |
Like MMS, we select `−0.2` as the default threshold and choose samples with scores greater than this threshold. We call the resultant subsets `clean`, which you can load as shown in the [Usage section](#usage).
|
| 40 |
|
| 41 |
-
> ⚠️ **2 Chronicles 13
|
| 42 |
|
| 43 |
We recommend reading the MMS paper for better understanding.
|
| 44 |
|
|
|
|
| 38 |
|
| 39 |
Like MMS, we select `−0.2` as the default threshold and choose samples with scores greater than this threshold. We call the resultant subsets `clean`, which you can load as shown in the [Usage section](#usage).
|
| 40 |
|
| 41 |
+
> ⚠️ **2 Chronicles 13-14 (`2CH_013`, `2CH_014`) have corrupt source audio. In the non-`clean` subsets, these chapters might be problematic for training!**
|
| 42 |
|
| 43 |
We recommend reading the MMS paper for better understanding.
|
| 44 |
|