Update readme for the Python Model
Browse filesHi,
I see that there is comment in the python module, where instead of transformer model the hybrid model is commented.
So just fixing the small issue.
Current
```
# model = Zonos.from_pretrained("Zyphra/Zonos-v0.1-hybrid", device="cuda")
model = Zonos.from_pretrained("Zyphra/Zonos-v0.1-transformer", device="cuda")
```
Changes to
```
model = Zonos.from_pretrained("Zyphra/Zonos-v0.1-hybrid", device="cuda")
# model = Zonos.from_pretrained("Zyphra/Zonos-v0.1-transformer", device="cuda")
```
README.md
CHANGED
@@ -46,8 +46,8 @@ import torchaudio
|
|
46 |
from zonos.model import Zonos
|
47 |
from zonos.conditioning import make_cond_dict
|
48 |
|
49 |
-
|
50 |
-
model = Zonos.from_pretrained("Zyphra/Zonos-v0.1-transformer", device="cuda")
|
51 |
|
52 |
wav, sampling_rate = torchaudio.load("assets/exampleaudio.mp3")
|
53 |
speaker = model.make_speaker_embedding(wav, sampling_rate)
|
|
|
46 |
from zonos.model import Zonos
|
47 |
from zonos.conditioning import make_cond_dict
|
48 |
|
49 |
+
model = Zonos.from_pretrained("Zyphra/Zonos-v0.1-hybrid", device="cuda")
|
50 |
+
# model = Zonos.from_pretrained("Zyphra/Zonos-v0.1-transformer", device="cuda")
|
51 |
|
52 |
wav, sampling_rate = torchaudio.load("assets/exampleaudio.mp3")
|
53 |
speaker = model.make_speaker_embedding(wav, sampling_rate)
|