Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
@@ -29,6 +29,13 @@ JS = """function () {
|
|
29 |
window.location.replace(gradioURL + '?__theme=dark');
|
30 |
}
|
31 |
}"""
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
32 |
|
33 |
# Ensure model and scheduler are initialized in GPU-enabled function
|
34 |
if torch.cuda.is_available():
|
@@ -70,7 +77,7 @@ def generate_image(
|
|
70 |
|
71 |
# Gradio Interface
|
72 |
|
73 |
-
with gr.Blocks(theme='soft', css=CSS, title="Stable Audio Open") as iface:
|
74 |
with gr.Accordion(""):
|
75 |
gr.Markdown(DESCRIPTION)
|
76 |
with gr.Row():
|
|
|
29 |
window.location.replace(gradioURL + '?__theme=dark');
|
30 |
}
|
31 |
}"""
|
32 |
+
DESCRIPTION = """
|
33 |
+
<center>
|
34 |
+
Stable Audio Open 1.0 generates variable-length (up to 47s) stereo audio at 44.1kHz from text prompts. \
|
35 |
+
It comprises three components: an autoencoder that compresses waveforms into a manageable sequence length, \
|
36 |
+
a T5-based text embedding for text conditioning, and a transformer-based diffusion (DiT) model that operates in the latent space of the autoencoder.
|
37 |
+
</center>
|
38 |
+
"""
|
39 |
|
40 |
# Ensure model and scheduler are initialized in GPU-enabled function
|
41 |
if torch.cuda.is_available():
|
|
|
77 |
|
78 |
# Gradio Interface
|
79 |
|
80 |
+
with gr.Blocks(theme='soft', css=CSS, js=JS, title="Stable Audio Open") as iface:
|
81 |
with gr.Accordion(""):
|
82 |
gr.Markdown(DESCRIPTION)
|
83 |
with gr.Row():
|