Spaces:
Running
Running
File size: 31,516 Bytes
3f56d56 2e667cb a677c6f 15f6d19 75d3f3c eb1c729 5cb13ea 1036aef d52c44f 1036aef 4788471 1036aef 2cc588e 37e4a75 1036aef 37e4a75 1036aef 2cc588e 1036aef 2e667cb 1036aef 2ac2781 1036aef 2ac2781 1036aef 2ac2781 1036aef 2ac2781 1036aef 2cc588e 2ac2781 1036aef 2ac2781 1036aef 2ac2781 1036aef 2ac2781 1036aef 2ac2781 1036aef 2ac2781 1036aef 2ac2781 1036aef 2cc588e 1036aef 2ac2781 1036aef 2ac2781 1036aef 2ac2781 1036aef 2ac2781 1036aef 2ac2781 1036aef 2cc588e 2ac2781 1036aef 2ac2781 1036aef 2ac2781 1036aef 2ac2781 1036aef 2ac2781 1036aef 2ac2781 1036aef 2ac2781 1036aef 2ac2781 1036aef 2cc588e 1036aef 2cc588e 1036aef 2ac2781 1036aef 2cc588e 1036aef 3f56d56 1036aef 2cc588e 1036aef 2cc588e 1036aef 2cc588e 1036aef ad9fa56 1036aef 824e8a4 1036aef ad9fa56 3a2a3a0 1036aef 3a2a3a0 1036aef 00523fc 24dac05 00523fc e42de32 ad9fa56 1036aef |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 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 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 |
import gradio as gr
from separwator import *
import os, subprocess
import gc, re
def leaderboard(list_filter, list_limit):
try:
result = subprocess.run(
["audio-separator", "-l", f"--list_filter={list_filter}", f"--list_limit={list_limit}"],
capture_output=True,
text=True,
)
if result.returncode != 0:
return f"Error: {result.stderr}"
return "<table border='1'>" + "".join(
f"<tr style='{'font-weight: bold; font-size: 1.2em;' if i == 0 else ''}'>" +
"".join(f"<td>{cell}</td>" for cell in re.split(r"\s{2,}", line.strip())) +
"</tr>"
for i, line in enumerate(re.findall(r"^(?!-+)(.+)$", result.stdout.strip(), re.MULTILINE))
) + "</table>"
except Exception as e:
return f"Error: {e}"
with gr.Blocks(theme=gr.themes.Base(), title = "🎵 Audio Separator UI 🎵") as app:
with gr.Row():
gr.Markdown("<h1><center> 🎵 Audio Separator UI 🎵")
with gr.Row():
with gr.Tabs():
with gr.TabItem("BS/Mel Roformer"):
with gr.Row():
roformer_model = gr.Dropdown(
label = "Select the model",
choices = list(roformer_models.keys()),
value = lambda : None,
interactive = True
)
roformer_output_format = gr.Dropdown(
label = "Select the output format",
choices = output_format,
value = lambda : None,
interactive = True
)
with gr.Row():
roformer_audio = gr.Audio(
label = "Input audio",
type = "filepath",
interactive = True
)
with gr.Column():
with gr.Accordion("Advanced settings", open = False):
with gr.Group():
with gr.Row():
roformer_segment_size = gr.Slider(
label = "Segment size",
info = "Larger consumes more resources, but may give better results",
minimum = 32,
maximum = 4000,
step = 32,
value = 256,
interactive = True
)
roformer_override_segment_size = gr.Checkbox(
label = "Override segment size",
info = "Override model default segment size instead of using the model default value",
value = False,
interactive = True
)
with gr.Row():
roformer_overlap = gr.Slider(
label = "Overlap",
info = "Amount of overlap between prediction windows",
minimum = 2,
maximum = 10,
step = 1,
value = 8,
interactive = True
)
roformer_batch_size = gr.Slider(
label = "Batch size",
info = "Larger consumes more RAM but may process slightly faster",
minimum = 1,
maximum = 16,
step = 1,
value = 1,
interactive = True
)
with gr.Row():
roformer_normalization_threshold = gr.Slider(
label = "Normalization threshold",
info = "The threshold for audio normalization",
minimum = 0.1,
maximum = 1,
step = 0.1,
value = 0.1,
interactive = True
)
roformer_amplification_threshold = gr.Slider(label = "Amplification threshold",info = "The threshold for audio amplification", minimum = 0.1,maximum = 1,step = 0.1,value = 0.1,interactive = True)
with gr.Row():
roformer_button = gr.Button("Separate!", variant = "primary")
with gr.Row():
roformer_stem1 = gr.Audio(
show_download_button = True,
interactive = False,
label = "Stem 1",
type = "filepath"
)
roformer_stem2 = gr.Audio(
show_download_button = True,
interactive = False,
label = "Stem 2",
type = "filepath"
)
roformer_button.click(roformer_separator, [roformer_audio, roformer_model, roformer_output_format, roformer_segment_size, roformer_override_segment_size, roformer_overlap, roformer_batch_size, roformer_normalization_threshold, roformer_amplification_threshold], [roformer_stem1, roformer_stem2])
with gr.TabItem("MDX23C"):
with gr.Row():
mdx23c_model = gr.Dropdown(
label = "Select the model",
choices = mdx23c_models,
value = lambda : None,
interactive = True
)
mdx23c_output_format = gr.Dropdown(
label = "Select the output format",
choices = output_format,
value = lambda : None,
interactive = True
)
with gr.Row():
mdx23c_audio = gr.Audio(
label = "Input audio",
type = "filepath",
interactive = True
)
with gr.Accordion("Advanced settings", open = False):
with gr.Group():
with gr.Row():
mdx23c_segment_size = gr.Slider(
minimum = 32,
maximum = 4000,
step = 32,
label = "Segment size",
info = "Larger consumes more resources, but may give better results",
value = 256,
interactive = True
)
mdx23c_override_segment_size = gr.Checkbox(
label = "Override segment size",
info = "Override model default segment size instead of using the model default value",
value = False,
interactive = True
)
with gr.Row():
mdx23c_overlap = gr.Slider(
minimum = 2,
maximum = 50,
step = 1,
label = "Overlap",
info = "Amount of overlap between prediction windows",
value = 8,
interactive = True
)
mdx23c_batch_size = gr.Slider(
label = "Batch size",
info = "Larger consumes more RAM but may process slightly faster",
minimum = 1,
maximum = 16,
step = 1,
value = 1,
interactive = True
)
with gr.Row():
mdx23c_normalization_threshold = gr.Slider(
label = "Normalization threshold",
info = "The threshold for audio normalization",
minimum = 0.1,
maximum = 1,
step = 0.1,
value = 0.1,
interactive = True
)
mdx23c_amplification_threshold = gr.Slider(
label = "Amplification threshold",
info = "The threshold for audio amplification",
minimum = 0.1,
maximum = 1,
step = 0.1,
value = 0.1,
interactive = True
)
with gr.Row():
mdx23c_button = gr.Button("Separate!", variant = "primary")
with gr.Row():
mdx23c_stem1 = gr.Audio(
show_download_button = True,
interactive = False,
label = "Stem 1",
type = "filepath"
)
mdx23c_stem2 = gr.Audio(
show_download_button = True,
interactive = False,
label = "Stem 2",
type = "filepath"
)
mdx23c_button.click(mdxc_separator, [mdx23c_audio, mdx23c_model, mdx23c_output_format, mdx23c_segment_size, mdx23c_override_segment_size, mdx23c_overlap, mdx23c_batch_size, mdx23c_normalization_threshold, mdx23c_amplification_threshold], [mdx23c_stem1, mdx23c_stem2])
with gr.TabItem("MDX-NET"):
with gr.Row():
mdxnet_model = gr.Dropdown(
label = "Select the model",
choices = mdxnet_models,
value = lambda : None,
interactive = True
)
mdxnet_output_format = gr.Dropdown(
label = "Select the output format",
choices = output_format,
value = lambda : None,
interactive = True
)
with gr.Row():
mdxnet_audio = gr.Audio(
label = "Input audio",
type = "filepath",
interactive = True
)
with gr.Accordion("Advanced settings", open = False):
with gr.Group():
with gr.Row():
mdxnet_hop_length = gr.Slider(
label = "Hop length",
info = "Usually called stride in neural networks; only change if you know what you're doing",
minimum = 32,
maximum = 2048,
step = 32,
value = 1024,
interactive = True
)
mdxnet_segment_size = gr.Slider(
minimum = 32,
maximum = 4000,
step = 32,
label = "Segment size",
info = "Larger consumes more resources, but may give better results",
value = 256,
interactive = True
)
mdxnet_denoise = gr.Checkbox(
label = "Denoise",
info = "Enable denoising during separation",
value = True,
interactive = True
)
with gr.Row():
mdxnet_overlap = gr.Slider(
label = "Overlap",
info = "Amount of overlap between prediction windows",
minimum = 0.001,
maximum = 0.999,
step = 0.001,
value = 0.25,
interactive = True
)
mdxnet_batch_size = gr.Slider(
label = "Batch size",
info = "Larger consumes more RAM but may process slightly faster",
minimum = 1,
maximum = 16,
step = 1,
value = 1,
interactive = True
)
with gr.Row():
mdxnet_normalization_threshold = gr.Slider(
label = "Normalization threshold",
info = "The threshold for audio normalization",
minimum = 0.1,
maximum = 1,
step = 0.1,
value = 0.1,
interactive = True
)
mdxnet_amplification_threshold = gr.Slider(
label = "Amplification threshold",
info = "The threshold for audio amplification",
minimum = 0.1,
maximum = 1,
step = 0.1,
value = 0.1,
interactive = True
)
with gr.Row():
mdxnet_button = gr.Button("Separate!", variant = "primary")
with gr.Row():
mdxnet_stem1 = gr.Audio(
show_download_button = True,
interactive = False,
label = "Stem 1",
type = "filepath"
)
mdxnet_stem2 = gr.Audio(
show_download_button = True,
interactive = False,
label = "Stem 2",
type = "filepath"
)
mdxnet_button.click(mdxnet_separator, [mdxnet_audio, mdxnet_model, mdxnet_output_format, mdxnet_hop_length, mdxnet_segment_size, mdxnet_denoise, mdxnet_overlap, mdxnet_batch_size, mdxnet_normalization_threshold, mdxnet_amplification_threshold], [mdxnet_stem1, mdxnet_stem2])
with gr.TabItem("VR ARCH"):
with gr.Row():
vrarch_model = gr.Dropdown(
label = "Select the model",
choices = vrarch_models,
value = lambda : None,
interactive = True
)
vrarch_output_format = gr.Dropdown(
label = "Select the output format",
choices = output_format,
value = lambda : None,
interactive = True
)
with gr.Row():
vrarch_audio = gr.Audio(
label = "Input audio",
type = "filepath",
interactive = True
)
with gr.Accordion("Advanced settings", open = False):
with gr.Group():
with gr.Row():
vrarch_window_size = gr.Slider(
label = "Window size",
info = "Balance quality and speed. 1024 = fast but lower, 320 = slower but better quality",
minimum=320,
maximum=1024,
step=32,
value = 512,
interactive = True
)
vrarch_agression = gr.Slider(
minimum = 1,
maximum = 50,
step = 1,
label = "Agression",
info = "Intensity of primary stem extraction",
value = 5,
interactive = True
)
vrarch_tta = gr.Checkbox(
label = "TTA",
info = "Enable Test-Time-Augmentation; slow but improves quality",
value = True,
visible = True,
interactive = True
)
with gr.Row():
vrarch_post_process = gr.Checkbox(
label = "Post process",
info = "Identify leftover artifacts within vocal output; may improve separation for some songs",
value = False,
visible = True,
interactive = True
)
vrarch_post_process_threshold = gr.Slider(
label = "Post process threshold",
info = "Threshold for post-processing",
minimum = 0.1,
maximum = 0.3,
step = 0.1,
value = 0.2,
interactive = True
)
with gr.Row():
vrarch_high_end_process = gr.Checkbox(
label = "High end process",
info = "Mirror the missing frequency range of the output",
value = False,
visible = True,
interactive = True,
)
vrarch_batch_size = gr.Slider(
label = "Batch size",
info = "Larger consumes more RAM but may process slightly faster",
minimum = 1,
maximum = 16,
step = 1,
value = 1,
interactive = True
)
with gr.Row():
vrarch_normalization_threshold = gr.Slider(
label = "Normalization threshold",
info = "The threshold for audio normalization",
minimum = 0.1,
maximum = 1,
step = 0.1,
value = 0.1,
interactive = True
)
vrarch_amplification_threshold = gr.Slider(
label = "Amplification threshold",
info = "The threshold for audio amplification",
minimum = 0.1,
maximum = 1,
step = 0.1,
value = 0.1,
interactive = True
)
with gr.Row():
vrarch_button = gr.Button("Separate!", variant = "primary")
with gr.Row():
vrarch_stem1 = gr.Audio(
show_download_button = True,
interactive = False,
type = "filepath",
label = "Stem 1"
)
vrarch_stem2 = gr.Audio(
show_download_button = True,
interactive = False,
type = "filepath",
label = "Stem 2"
)
vrarch_button.click(vrarch_separator, [vrarch_audio, vrarch_model, vrarch_output_format, vrarch_window_size, vrarch_agression, vrarch_tta, vrarch_post_process, vrarch_post_process_threshold, vrarch_high_end_process, vrarch_batch_size, vrarch_normalization_threshold, vrarch_amplification_threshold], [vrarch_stem1, vrarch_stem2])
with gr.TabItem("Demucs"):
with gr.Row():
demucs_model = gr.Dropdown(
label = "Select the model",
choices = demucs_models,
value = lambda : None,
interactive = True
)
demucs_output_format = gr.Dropdown(
label = "Select the output format",
choices = output_format,
value = lambda : None,
interactive = True
)
with gr.Row():
demucs_audio = gr.Audio(
label = "Input audio",
type = "filepath",
interactive = True
)
with gr.Accordion("Advanced settings", open = False):
with gr.Group():
with gr.Row():
demucs_shifts = gr.Slider(
label = "Shifts",
info = "Number of predictions with random shifts, higher = slower but better quality",
minimum = 1,
maximum = 20,
step = 1,
value = 2,
interactive = True
)
demucs_segment_size = gr.Slider(
label = "Segment size",
info = "Size of segments into which the audio is split. Higher = slower but better quality",
minimum = 1,
maximum = 100,
step = 1,
value = 40,
interactive = True
)
demucs_segments_enabled = gr.Checkbox(
label = "Segment-wise processing",
info = "Enable segment-wise processing",
value = True,
interactive = True
)
with gr.Row():
demucs_overlap = gr.Slider(
label = "Overlap",
info = "Overlap between prediction windows. Higher = slower but better quality",
minimum=0.001,
maximum=0.999,
step=0.001,
value = 0.25,
interactive = True
)
demucs_batch_size = gr.Slider(
label = "Batch size",
info = "Larger consumes more RAM but may process slightly faster",
minimum = 1,
maximum = 16,
step = 1,
value = 1,
interactive = True
)
with gr.Row():
demucs_normalization_threshold = gr.Slider(
label = "Normalization threshold",
info = "The threshold for audio normalization",
minimum = 0.1,
maximum = 1,
step = 0.1,
value = 0.1,
interactive = True
)
demucs_amplification_threshold = gr.Slider(
label = "Amplification threshold",
info = "The threshold for audio amplification",
minimum = 0.1,
maximum = 1,
step = 0.1,
value = 0.1,
interactive = True
)
with gr.Row():
gr.Markdown("You can paste the link to the video/audio from many sites, check the complete list [here](https://github.com/yt-dlp/yt-dlp/blob/master/supportedsites.md)")
with gr.Row():
demucs_download_button = gr.Button(
"Download!",
variant = "primary"
)
with gr.Row():
demucs_button = gr.Button("Separate!", variant = "primary")
with gr.Row():
demucs_stem1 = gr.Audio(
show_download_button = True,
interactive = False,
type = "filepath",
label = "Stem 1"
)
demucs_stem2 = gr.Audio(
show_download_button = True,
interactive = False,
type = "filepath",
label = "Stem 2"
)
with gr.Row():
demucs_stem3 = gr.Audio(
show_download_button = True,
interactive = False,
type = "filepath",
label = "Stem 3"
)
demucs_stem4 = gr.Audio(
show_download_button = True,
interactive = False,
type = "filepath",
label = "Stem 4"
)
with gr.Row(visible=False) as stem6:
demucs_stem5 = gr.Audio(
show_download_button = True,
interactive = False,
type = "filepath",
label = "Stem 5"
)
demucs_stem6 = gr.Audio(
show_download_button = True,
interactive = False,
type = "filepath",
label = "Stem 6"
)
demucs_model.change(update_stems, inputs=[demucs_model], outputs=stem6)
demucs_button.click(demucs_separator, [demucs_audio, demucs_model, demucs_output_format, demucs_shifts, demucs_segment_size, demucs_segments_enabled, demucs_overlap, demucs_batch_size, demucs_normalization_threshold, demucs_amplification_threshold], [demucs_stem1, demucs_stem2, demucs_stem3, demucs_stem4, demucs_stem5, demucs_stem6])
with gr.Tab("Leaderboard"):
with gr.Row(equal_height=True):
list_filter = gr.Dropdown(value="vocals", choices=["vocals", "instrumental", "drums", "bass", "guitar", "piano", "other"], label="List filter", info="Filter and sort the model list by 'stem'")
list_limit = gr.Slider(minimum=1, maximum=10, step=1, value=5, label="List limit", info="Limit the number of models shown.")
list_button = gr.Button("Show list", variant="primary")
output_list = gr.HTML(label="Leaderboard")
list_button.click(leaderboard, inputs=[list_filter, list_limit], outputs=output_list)
with gr.TabItem("Credits"):
gr.Markdown(
"""
audio separator UI created by [Eddycrack 864](https://github.com/Eddycrack864) & [_noxty](https://huggingface.co/theNeofr).
* python-audio-separator by [beveradb](https://github.com/beveradb).
* Thanks to [Mikus](https://github.com/cappuch) for the help with the code.
* Thanks to [Nick088](https://huggingface.co/Nick088) for the help to fix roformers.
* Thanks to [ArisDev](https://github.com/aris-py) for porting UVR5 UI to Kaggle and improvements.
* Thanks to [Bebra777228](https://github.com/Bebra777228)'s code for guiding me to improve my code.
You can donate to the original UVR5 project here:
[](https://www.buymeacoffee.com/uvr5)
"""
)
gr.Markdown(
"""
<h1> similar of this project\n
[Audio_separator by r3gm](https://huggingface.co/spaces/r3gm/Audio_separator)\n
[Audio-Separator by Politrees](https://huggingface.co/spaces/Politrees/Audio-Separator)\n
[UVR5 UI by Eddycrack 864](https://huggingface.co/spaces/Eddycrack864/UVR5-UI)
"""
)
app.queue()
app.launch(share=True, debug=True) |