huseinzol05 commited on
Commit
d47b265
1 Parent(s): 3929881

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +35 -0
README.md ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ datasets:
3
+ - mesolitica/Malaysian-Emilia
4
+ language:
5
+ - ms
6
+ - en
7
+ base_model:
8
+ - charactr/vocos-mel-24khz
9
+ ---
10
+ # Malaysian Vocos
11
+
12
+ Continue pretraining [charactr/vocos-mel-24khz](https://huggingface.co/charactr/vocos-mel-24khz) on [Malaysian Emilia](https://huggingface.co/datasets/mesolitica/Malaysian-Emilia), to make it more crispy!
13
+
14
+ ## Installation
15
+
16
+ To use Vocos only in inference mode, install it using:
17
+
18
+ ```bash
19
+ pip install vocos
20
+ ```
21
+
22
+ ## Usage
23
+
24
+ ### Reconstruct audio from mel-spectrogram
25
+
26
+ ```python
27
+ import torch
28
+
29
+ from vocos import Vocos
30
+
31
+ vocos = Vocos.from_pretrained("mesolitica/malaysian-vocos-mel-24khz")
32
+
33
+ mel = torch.randn(1, 100, 256) # B, C, T
34
+ audio = vocos.decode(mel)
35
+ ```