ameerazam08 commited on
Commit
35fe8b4
1 Parent(s): ccf495a

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +56 -1
app.py CHANGED
@@ -98,7 +98,62 @@ interface = gr.Interface(
98
  ],
99
  outputs=gr.Audio(type="filepath", label="Generated Audio"),
100
  title="Stable Audio Generator",
101
- description="Generate variable-length stereo audio at 44.1kHz from text prompts using Stable Audio Open 1.0."
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
102
  )
103
 
104
  # Pre-load the model to avoid multiprocessing issues
 
98
  ],
99
  outputs=gr.Audio(type="filepath", label="Generated Audio"),
100
  title="Stable Audio Generator",
101
+ description="Generate variable-length stereo audio at 44.1kHz from text prompts using Stable Audio Open 1.0.",
102
+ examples=[
103
+ [
104
+ "Create a serene soundscape of a quiet beach at sunset.", # Text prompt
105
+ "dpmpp-2m-sde", # Sampler type
106
+ 45, # Duration in Seconds
107
+ 100, # Number of Diffusion Steps
108
+ 10, # CFG Scale
109
+ 0.5, # Sigma min
110
+ 800 # Sigma max
111
+ ],
112
+ [
113
+ "Generate an energetic and bustling city street scene with distant traffic and close conversations.", # Text prompt
114
+ "k-heun", # Sampler type
115
+ 30, # Duration in Seconds
116
+ 120, # Number of Diffusion Steps
117
+ 5, # CFG Scale
118
+ 0.2, # Sigma min
119
+ 600 # Sigma max
120
+ ],
121
+ [
122
+ "Simulate a forest ambiance with birds chirping and wind rustling through the leaves.", # Text prompt
123
+ "k-dpm-fast", # Sampler type
124
+ 60, # Duration in Seconds
125
+ 140, # Number of Diffusion Steps
126
+ 7.5, # CFG Scale
127
+ 0.3, # Sigma min
128
+ 700 # Sigma max
129
+ ],
130
+ [
131
+ "Recreate a gentle rainfall with distant thunder.", # Text prompt
132
+ "dpmpp-3m-sde", # Sampler type
133
+ 35, # Duration in Seconds
134
+ 110, # Number of Diffusion Steps
135
+ 8, # CFG Scale
136
+ 0.1, # Sigma min
137
+ 500 # Sigma max
138
+ ],
139
+ [
140
+ "Imagine a jazz cafe environment with soft music and ambient chatter.", # Text prompt
141
+ "k-lms", # Sampler type
142
+ 25, # Duration in Seconds
143
+ 90, # Number of Diffusion Steps
144
+ 6, # CFG Scale
145
+ 0.4, # Sigma min
146
+ 650 # Sigma max
147
+ ],
148
+ ["Rock beat played in a treated studio, session drumming on an acoustic kit.",
149
+ "k-lms", # Sampler type
150
+ 25, # Duration in Seconds
151
+ 90, # Number of Diffusion Steps
152
+ 6, # CFG Scale
153
+ 0.3, # Sigma min
154
+ 500 # Sigma max
155
+ ]
156
+ ]
157
  )
158
 
159
  # Pre-load the model to avoid multiprocessing issues