Spaces:
Running
Running
flosstradamus
commited on
Update app.py
Browse files
app.py
CHANGED
@@ -5,15 +5,10 @@ from einops import rearrange, repeat
|
|
5 |
from diffusers import AutoencoderKL
|
6 |
from transformers import SpeechT5HifiGan
|
7 |
from scipy.io import wavfile
|
|
|
8 |
import random
|
9 |
import numpy as np
|
10 |
import re
|
11 |
-
import requests
|
12 |
-
from urllib.parse import urlparse
|
13 |
-
import logging
|
14 |
-
|
15 |
-
# Set up logging
|
16 |
-
logging.basicConfig(level=logging.INFO)
|
17 |
|
18 |
# Import necessary functions and classes
|
19 |
from utils import load_t5, load_clap
|
@@ -28,12 +23,44 @@ global_vae = None
|
|
28 |
global_vocoder = None
|
29 |
global_diffusion = None
|
30 |
|
31 |
-
# Set the
|
|
|
32 |
GENERATIONS_DIR = "/content/generations"
|
33 |
|
34 |
def prepare(t5, clip, img, prompt):
|
35 |
-
|
36 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
37 |
|
38 |
def unload_current_model():
|
39 |
global global_model
|
@@ -42,171 +69,152 @@ def unload_current_model():
|
|
42 |
torch.cuda.empty_cache()
|
43 |
global_model = None
|
44 |
|
45 |
-
def
|
46 |
-
try:
|
47 |
-
response = requests.get(url, stream=True)
|
48 |
-
if response.status_code == 200:
|
49 |
-
filename = os.path.basename(urlparse(url).path)
|
50 |
-
model_path = os.path.join("/tmp", filename)
|
51 |
-
with open(model_path, "wb") as f:
|
52 |
-
for chunk in response.iter_content(chunk_size=8192):
|
53 |
-
f.write(chunk)
|
54 |
-
return model_path
|
55 |
-
else:
|
56 |
-
raise Exception(f"Failed to download model from {url}")
|
57 |
-
except Exception as e:
|
58 |
-
logging.error(f"Error downloading model: {str(e)}")
|
59 |
-
raise
|
60 |
-
|
61 |
-
def load_model(url):
|
62 |
global global_model
|
63 |
-
|
64 |
-
|
65 |
-
|
66 |
-
|
67 |
-
|
68 |
-
|
69 |
-
|
70 |
-
|
71 |
-
|
72 |
-
|
73 |
-
|
74 |
-
|
75 |
-
|
76 |
-
|
77 |
-
|
78 |
-
|
79 |
-
|
80 |
-
|
81 |
-
|
82 |
-
|
83 |
-
|
84 |
-
|
85 |
-
|
86 |
-
|
87 |
-
state_dict = torch.load(model_path, map_location=lambda storage, loc: storage, weights_only=True)
|
88 |
-
global_model.load_state_dict(state_dict['ema'])
|
89 |
-
global_model.eval()
|
90 |
-
global_model.model_path = model_path
|
91 |
-
logging.info("Model loaded successfully")
|
92 |
-
return "Model loaded successfully"
|
93 |
-
except Exception as e:
|
94 |
-
logging.error(f"Error loading model: {str(e)}")
|
95 |
-
return f"Error loading model: {str(e)}"
|
96 |
|
97 |
def load_resources():
|
98 |
global global_t5, global_clap, global_vae, global_vocoder, global_diffusion
|
99 |
|
100 |
device = "cuda" if torch.cuda.is_available() else "cpu"
|
101 |
|
102 |
-
|
103 |
global_t5 = load_t5(device, max_length=256)
|
104 |
global_clap = load_clap(device, max_length=256)
|
105 |
|
106 |
-
|
107 |
global_vae = AutoencoderKL.from_pretrained('cvssp/audioldm2', subfolder="vae").to(device)
|
108 |
global_vocoder = SpeechT5HifiGan.from_pretrained('cvssp/audioldm2', subfolder="vocoder").to(device)
|
109 |
|
110 |
-
|
111 |
global_diffusion = RF()
|
112 |
|
113 |
-
|
114 |
|
115 |
-
def generate_music(prompt, seed, cfg_scale, steps, duration):
|
116 |
global global_model, global_t5, global_clap, global_vae, global_vocoder, global_diffusion
|
117 |
|
118 |
if global_model is None:
|
119 |
-
return "Please
|
120 |
|
121 |
if seed == 0:
|
122 |
seed = random.randint(1, 1000000)
|
123 |
-
|
124 |
|
125 |
device = "cuda" if torch.cuda.is_available() else "cpu"
|
126 |
torch.manual_seed(seed)
|
127 |
torch.set_grad_enabled(False)
|
128 |
|
129 |
-
|
130 |
-
|
131 |
-
|
132 |
-
num_segments = int(np.ceil(duration / segment_duration))
|
133 |
|
134 |
-
|
135 |
|
136 |
-
|
137 |
-
|
138 |
|
139 |
-
|
140 |
-
|
141 |
|
142 |
-
|
143 |
-
|
144 |
-
|
145 |
-
|
146 |
|
147 |
-
|
148 |
|
149 |
-
|
150 |
-
|
151 |
|
152 |
-
|
153 |
-
|
154 |
|
155 |
-
|
156 |
-
|
157 |
-
|
158 |
-
|
159 |
-
|
160 |
-
|
161 |
-
|
162 |
|
163 |
-
|
164 |
-
|
165 |
|
166 |
-
|
167 |
-
|
168 |
-
|
169 |
-
|
170 |
-
|
171 |
|
172 |
-
|
173 |
|
174 |
-
|
175 |
-
|
176 |
|
177 |
-
|
178 |
-
|
179 |
-
|
180 |
|
181 |
-
|
182 |
-
|
183 |
-
|
184 |
-
|
185 |
-
|
186 |
-
|
187 |
-
|
188 |
-
|
189 |
-
|
190 |
-
|
191 |
-
|
192 |
-
|
193 |
-
|
194 |
-
|
195 |
-
|
196 |
-
|
197 |
-
|
198 |
|
199 |
-
|
200 |
|
201 |
-
|
202 |
-
|
203 |
-
except Exception as e:
|
204 |
-
logging.error(f"Error generating music: {str(e)}")
|
205 |
-
return f"Error generating music: {str(e)}", None
|
206 |
|
207 |
# Load base resources at startup
|
208 |
load_resources()
|
209 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
210 |
# Set up dark grey theme
|
211 |
theme = gr.themes.Monochrome(
|
212 |
primary_hue="gray",
|
@@ -225,29 +233,32 @@ with gr.Blocks(theme=theme) as iface:
|
|
225 |
</div>
|
226 |
""")
|
227 |
|
228 |
-
|
229 |
-
|
230 |
-
|
|
|
|
|
|
|
231 |
|
232 |
-
|
233 |
-
|
234 |
-
|
235 |
-
|
236 |
-
duration = gr.Number(label="Duration (seconds)", value=10, minimum=10, maximum=300, step=1)
|
237 |
|
238 |
generate_button = gr.Button("Generate Music")
|
239 |
output_status = gr.Textbox(label="Generation Status")
|
240 |
output_audio = gr.Audio(type="filepath")
|
241 |
|
242 |
-
def
|
243 |
-
|
244 |
|
245 |
-
|
246 |
-
|
247 |
-
return status, audio_path if audio_path else None
|
248 |
|
249 |
-
|
250 |
-
|
|
|
|
|
251 |
|
252 |
# Launch the interface
|
253 |
iface.launch()
|
|
|
5 |
from diffusers import AutoencoderKL
|
6 |
from transformers import SpeechT5HifiGan
|
7 |
from scipy.io import wavfile
|
8 |
+
import glob
|
9 |
import random
|
10 |
import numpy as np
|
11 |
import re
|
|
|
|
|
|
|
|
|
|
|
|
|
12 |
|
13 |
# Import necessary functions and classes
|
14 |
from utils import load_t5, load_clap
|
|
|
23 |
global_vocoder = None
|
24 |
global_diffusion = None
|
25 |
|
26 |
+
# Set the models directory
|
27 |
+
MODELS_DIR = "/content/models"
|
28 |
GENERATIONS_DIR = "/content/generations"
|
29 |
|
30 |
def prepare(t5, clip, img, prompt):
|
31 |
+
bs, c, h, w = img.shape
|
32 |
+
if bs == 1 and not isinstance(prompt, str):
|
33 |
+
bs = len(prompt)
|
34 |
+
|
35 |
+
img = rearrange(img, "b c (h ph) (w pw) -> b (h w) (c ph pw)", ph=2, pw=2)
|
36 |
+
if img.shape[0] == 1 and bs > 1:
|
37 |
+
img = repeat(img, "1 ... -> bs ...", bs=bs)
|
38 |
+
|
39 |
+
img_ids = torch.zeros(h // 2, w // 2, 3)
|
40 |
+
img_ids[..., 1] = img_ids[..., 1] + torch.arange(h // 2)[:, None]
|
41 |
+
img_ids[..., 2] = img_ids[..., 2] + torch.arange(w // 2)[None, :]
|
42 |
+
img_ids = repeat(img_ids, "h w c -> b (h w) c", b=bs)
|
43 |
+
|
44 |
+
if isinstance(prompt, str):
|
45 |
+
prompt = [prompt]
|
46 |
+
|
47 |
+
# Generate text embeddings
|
48 |
+
txt = t5(prompt)
|
49 |
+
|
50 |
+
if txt.shape[0] == 1 and bs > 1:
|
51 |
+
txt = repeat(txt, "1 ... -> bs ...", bs=bs)
|
52 |
+
txt_ids = torch.zeros(bs, txt.shape[1], 3)
|
53 |
+
|
54 |
+
vec = clip(prompt)
|
55 |
+
if vec.shape[0] == 1 and bs > 1:
|
56 |
+
vec = repeat(vec, "1 ... -> bs ...", bs=bs)
|
57 |
+
|
58 |
+
return img, {
|
59 |
+
"img_ids": img_ids.to(img.device),
|
60 |
+
"txt": txt.to(img.device),
|
61 |
+
"txt_ids": txt_ids.to(img.device),
|
62 |
+
"y": vec.to(img.device),
|
63 |
+
}
|
64 |
|
65 |
def unload_current_model():
|
66 |
global global_model
|
|
|
69 |
torch.cuda.empty_cache()
|
70 |
global_model = None
|
71 |
|
72 |
+
def load_model(model_name):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
73 |
global global_model
|
74 |
+
device = "cuda" if torch.cuda.is_available() else "cpu"
|
75 |
+
|
76 |
+
unload_current_model()
|
77 |
+
|
78 |
+
# Determine model size from filename
|
79 |
+
if 'musicflow_b' in model_name:
|
80 |
+
model_size = "base"
|
81 |
+
elif 'musicflow_g' in model_name:
|
82 |
+
model_size = "giant"
|
83 |
+
elif 'musicflow_l' in model_name:
|
84 |
+
model_size = "large"
|
85 |
+
elif 'musicflow_s' in model_name:
|
86 |
+
model_size = "small"
|
87 |
+
else:
|
88 |
+
model_size = "base" # Default to base if unrecognized
|
89 |
+
|
90 |
+
print(f"Loading {model_size} model: {model_name}")
|
91 |
+
|
92 |
+
model_path = os.path.join(MODELS_DIR, model_name)
|
93 |
+
global_model = build_model(model_size).to(device)
|
94 |
+
state_dict = torch.load(model_path, map_location=lambda storage, loc: storage, weights_only=True)
|
95 |
+
global_model.load_state_dict(state_dict['ema'])
|
96 |
+
global_model.eval()
|
97 |
+
global_model.model_path = model_path
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
98 |
|
99 |
def load_resources():
|
100 |
global global_t5, global_clap, global_vae, global_vocoder, global_diffusion
|
101 |
|
102 |
device = "cuda" if torch.cuda.is_available() else "cpu"
|
103 |
|
104 |
+
print("Loading T5 and CLAP models...")
|
105 |
global_t5 = load_t5(device, max_length=256)
|
106 |
global_clap = load_clap(device, max_length=256)
|
107 |
|
108 |
+
print("Loading VAE and vocoder...")
|
109 |
global_vae = AutoencoderKL.from_pretrained('cvssp/audioldm2', subfolder="vae").to(device)
|
110 |
global_vocoder = SpeechT5HifiGan.from_pretrained('cvssp/audioldm2', subfolder="vocoder").to(device)
|
111 |
|
112 |
+
print("Initializing diffusion...")
|
113 |
global_diffusion = RF()
|
114 |
|
115 |
+
print("Base resources loaded successfully!")
|
116 |
|
117 |
+
def generate_music(prompt, seed, cfg_scale, steps, duration, progress=gr.Progress()):
|
118 |
global global_model, global_t5, global_clap, global_vae, global_vocoder, global_diffusion
|
119 |
|
120 |
if global_model is None:
|
121 |
+
return "Please select a model first.", None
|
122 |
|
123 |
if seed == 0:
|
124 |
seed = random.randint(1, 1000000)
|
125 |
+
print(f"Using seed: {seed}")
|
126 |
|
127 |
device = "cuda" if torch.cuda.is_available() else "cpu"
|
128 |
torch.manual_seed(seed)
|
129 |
torch.set_grad_enabled(False)
|
130 |
|
131 |
+
# Calculate the number of segments needed for the desired duration
|
132 |
+
segment_duration = 10 # Each segment is 10 seconds
|
133 |
+
num_segments = int(np.ceil(duration / segment_duration))
|
|
|
134 |
|
135 |
+
all_waveforms = []
|
136 |
|
137 |
+
for i in range(num_segments):
|
138 |
+
progress(i / num_segments, desc=f"Generating segment {i+1}/{num_segments}")
|
139 |
|
140 |
+
# Use the same seed for all segments
|
141 |
+
torch.manual_seed(seed + i) # Add i to slightly vary each segment while maintaining consistency
|
142 |
|
143 |
+
latent_size = (256, 16)
|
144 |
+
conds_txt = [prompt]
|
145 |
+
unconds_txt = ["low quality, gentle"]
|
146 |
+
L = len(conds_txt)
|
147 |
|
148 |
+
init_noise = torch.randn(L, 8, latent_size[0], latent_size[1]).to(device)
|
149 |
|
150 |
+
img, conds = prepare(global_t5, global_clap, init_noise, conds_txt)
|
151 |
+
_, unconds = prepare(global_t5, global_clap, init_noise, unconds_txt)
|
152 |
|
153 |
+
with torch.autocast(device_type='cuda'):
|
154 |
+
images = global_diffusion.sample_with_xps(global_model, img, conds=conds, null_cond=unconds, sample_steps=steps, cfg=cfg_scale)
|
155 |
|
156 |
+
images = rearrange(
|
157 |
+
images[-1],
|
158 |
+
"b (h w) (c ph pw) -> b c (h ph) (w pw)",
|
159 |
+
h=128,
|
160 |
+
w=8,
|
161 |
+
ph=2,
|
162 |
+
pw=2,)
|
163 |
|
164 |
+
latents = 1 / global_vae.config.scaling_factor * images
|
165 |
+
mel_spectrogram = global_vae.decode(latents).sample
|
166 |
|
167 |
+
x_i = mel_spectrogram[0]
|
168 |
+
if x_i.dim() == 4:
|
169 |
+
x_i = x_i.squeeze(1)
|
170 |
+
waveform = global_vocoder(x_i)
|
171 |
+
waveform = waveform[0].cpu().float().detach().numpy()
|
172 |
|
173 |
+
all_waveforms.append(waveform)
|
174 |
|
175 |
+
# Concatenate all waveforms
|
176 |
+
final_waveform = np.concatenate(all_waveforms)
|
177 |
|
178 |
+
# Trim to exact duration
|
179 |
+
sample_rate = 16000
|
180 |
+
final_waveform = final_waveform[:int(duration * sample_rate)]
|
181 |
|
182 |
+
progress(0.9, desc="Saving audio file")
|
183 |
+
|
184 |
+
# Create 'generations' folder
|
185 |
+
os.makedirs(GENERATIONS_DIR, exist_ok=True)
|
186 |
+
|
187 |
+
# Generate filename
|
188 |
+
prompt_part = re.sub(r'[^\w\s-]', '', prompt)[:10].strip().replace(' ', '_')
|
189 |
+
model_name = os.path.splitext(os.path.basename(global_model.model_path))[0]
|
190 |
+
model_suffix = '_mf_b' if model_name == 'musicflow_b' else f'_{model_name}'
|
191 |
+
base_filename = f"{prompt_part}_{seed}{model_suffix}"
|
192 |
+
output_path = os.path.join(GENERATIONS_DIR, f"{base_filename}.wav")
|
193 |
+
|
194 |
+
# Check if file exists and add numerical suffix if needed
|
195 |
+
counter = 1
|
196 |
+
while os.path.exists(output_path):
|
197 |
+
output_path = os.path.join(GENERATIONS_DIR, f"{base_filename}_{counter}.wav")
|
198 |
+
counter += 1
|
199 |
|
200 |
+
wavfile.write(output_path, sample_rate, final_waveform)
|
201 |
|
202 |
+
progress(1.0, desc="Audio generation complete")
|
203 |
+
return f"Generated with seed: {seed}", output_path
|
|
|
|
|
|
|
204 |
|
205 |
# Load base resources at startup
|
206 |
load_resources()
|
207 |
|
208 |
+
# Get list of .pt files in the models directory
|
209 |
+
model_files = glob.glob(os.path.join(MODELS_DIR, "*.pt"))
|
210 |
+
model_choices = [os.path.basename(f) for f in model_files]
|
211 |
+
|
212 |
+
# Ensure 'musicflow_b.pt' is the default choice if it exists
|
213 |
+
default_model = 'musicflow_b.pt'
|
214 |
+
if default_model in model_choices:
|
215 |
+
model_choices.remove(default_model)
|
216 |
+
model_choices.insert(0, default_model)
|
217 |
+
|
218 |
# Set up dark grey theme
|
219 |
theme = gr.themes.Monochrome(
|
220 |
primary_hue="gray",
|
|
|
233 |
</div>
|
234 |
""")
|
235 |
|
236 |
+
with gr.Row():
|
237 |
+
model_dropdown = gr.Dropdown(choices=model_choices, label="Select Model", value=default_model if default_model in model_choices else model_choices[0])
|
238 |
+
|
239 |
+
with gr.Row():
|
240 |
+
prompt = gr.Textbox(label="Prompt")
|
241 |
+
seed = gr.Number(label="Seed", value=0)
|
242 |
|
243 |
+
with gr.Row():
|
244 |
+
cfg_scale = gr.Slider(minimum=1, maximum=40, step=0.1, label="CFG Scale", value=20)
|
245 |
+
steps = gr.Slider(minimum=10, maximum=200, step=1, label="Steps", value=100)
|
246 |
+
duration = gr.Number(label="Duration (seconds)", value=10, minimum=10, maximum=300, step=1)
|
|
|
247 |
|
248 |
generate_button = gr.Button("Generate Music")
|
249 |
output_status = gr.Textbox(label="Generation Status")
|
250 |
output_audio = gr.Audio(type="filepath")
|
251 |
|
252 |
+
def on_model_change(model_name):
|
253 |
+
load_model(model_name)
|
254 |
|
255 |
+
model_dropdown.change(on_model_change, inputs=[model_dropdown])
|
256 |
+
generate_button.click(generate_music, inputs=[prompt, seed, cfg_scale, steps, duration], outputs=[output_status, output_audio])
|
|
|
257 |
|
258 |
+
# Load default model on startup
|
259 |
+
default_model_path = os.path.join(MODELS_DIR, default_model)
|
260 |
+
if os.path.exists(default_model_path):
|
261 |
+
iface.load(lambda: load_model(default_model), inputs=None, outputs=None)
|
262 |
|
263 |
# Launch the interface
|
264 |
iface.launch()
|