update readme
Browse files
README.md
CHANGED
@@ -1,3 +1,41 @@
|
|
1 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2 |
license: cc-by-4.0
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
+
tags:
|
3 |
+
- espnet
|
4 |
+
- audio
|
5 |
+
- audio-to-audio
|
6 |
+
- vocoder
|
7 |
+
language:
|
8 |
+
- en
|
9 |
+
datasets:
|
10 |
+
- vctk
|
11 |
license: cc-by-4.0
|
12 |
+
inference: false
|
13 |
---
|
14 |
+
|
15 |
+
## Vocoder model - HifiGAN - English
|
16 |
+
|
17 |
+
https://github.com/kan-bayashi/ParallelWaveGAN
|
18 |
+
|
19 |
+
**No support given.**
|
20 |
+
|
21 |
+
### Details
|
22 |
+
|
23 |
+
```
|
24 |
+
sampling_rate: 44100 # Sampling rate.
|
25 |
+
fft_size: 2048 # FFT size.
|
26 |
+
hop_size: 512 # Hop size.
|
27 |
+
win_length: 2048 # Window length.
|
28 |
+
# If set to null, it will be the same as fft_size.
|
29 |
+
window: "hann" # Window function.
|
30 |
+
num_mels: 80 # Number of mel basis.
|
31 |
+
fmin: 0 # Minimum freq in mel basis calculation.
|
32 |
+
fmax: 22050 # Maximum frequency in mel basis calculation.
|
33 |
+
generator_type: HiFiGANGenerator
|
34 |
+
generator_params:
|
35 |
+
in_channels: 80 # Number of input channels.
|
36 |
+
out_channels: 1 # Number of output channels.
|
37 |
+
channels: 512 # Number of initial channels.
|
38 |
+
kernel_size: 7 # Kernel size of initial and final conv layers.
|
39 |
+
upsample_scales: [8, 8, 2, 2, 2] # Upsampling scales.
|
40 |
+
upsample_kernel_sizes: [16, 16, 4, 4, 4] # Kernel size for upsampling layers.
|
41 |
+
```
|