ahnafsamin
commited on
Commit
•
b57e9c4
1
Parent(s):
bc29aa3
Update README.md
Browse files
README.md
CHANGED
@@ -10,7 +10,7 @@ license: afl-3.0
|
|
10 |
---
|
11 |
## GroTTS Model
|
12 |
|
13 |
-
This model was trained with the [FastSpeech 2](https://arxiv.org/abs/2006.04558) architecture using approx. 2 hours of Gronings TTS dataset. For the best results you need to download the vocoder separately from [here](https://huggingface.co/ahnafsamin/parallelwavegan-gronings) and then use the following code:
|
14 |
```
|
15 |
from espnet2.bin.tts_inference import Text2Speech
|
16 |
from scipy.io.wavfile import write
|
@@ -23,6 +23,8 @@ output = model("This is a simple test.")
|
|
23 |
write("x.wav", 22050, output['wav'].numpy())
|
24 |
```
|
25 |
|
|
|
|
|
26 |
## TTS config
|
27 |
|
28 |
<details><summary>expand</summary>
|
|
|
10 |
---
|
11 |
## GroTTS Model
|
12 |
|
13 |
+
This model was trained with the [FastSpeech 2](https://arxiv.org/abs/2006.04558) architecture using approx. 2 hours of Gronings TTS dataset. For the best results, you need to download the vocoder separately from [here](https://huggingface.co/ahnafsamin/parallelwavegan-gronings) and then use the following code:
|
14 |
```
|
15 |
from espnet2.bin.tts_inference import Text2Speech
|
16 |
from scipy.io.wavfile import write
|
|
|
23 |
write("x.wav", 22050, output['wav'].numpy())
|
24 |
```
|
25 |
|
26 |
+
The GroTTS model is deployed [here](https://huggingface.co/spaces/ahnafsamin/GroTTS-FastSpeech2).
|
27 |
+
|
28 |
## TTS config
|
29 |
|
30 |
<details><summary>expand</summary>
|