File size: 23,436 Bytes
85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1ce1196 1a7d583 1ce1196 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 037701f 1a7d583 037701f 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 037701f 1a7d583 85d3b29 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 |
import os, sys
import gradio as gr
import regex as re
import shutil
import datetime
import random
from core import (
run_infer_script,
run_batch_infer_script,
)
from assets.i18n.i18n import I18nAuto
from rvc.lib.utils import format_title
i18n = I18nAuto()
now_dir = os.getcwd()
sys.path.append(now_dir)
model_root = os.path.join(now_dir, "logs")
audio_root = os.path.join(now_dir, "assets", "audios")
model_root_relative = os.path.relpath(model_root, now_dir)
audio_root_relative = os.path.relpath(audio_root, now_dir)
sup_audioext = {
"wav",
"mp3",
"flac",
"ogg",
"opus",
"m4a",
"mp4",
"aac",
"alac",
"wma",
"aiff",
"webm",
"ac3",
}
names = [
os.path.join(root, file)
for root, _, files in os.walk(model_root_relative, topdown=False)
for file in files
if (
file.endswith((".pth", ".onnx"))
and not (file.startswith("G_") or file.startswith("D_"))
)
]
indexes_list = [
os.path.join(root, name)
for root, _, files in os.walk(model_root_relative, topdown=False)
for name in files
if name.endswith(".index") and "trained" not in name
]
audio_paths = [
os.path.join(root, name)
for root, _, files in os.walk(audio_root_relative, topdown=False)
for name in files
if name.endswith(tuple(sup_audioext))
and root == audio_root_relative
and "_output" not in name
]
def output_path_fn(input_audio_path):
original_name_without_extension = os.path.basename(input_audio_path).rsplit(".", 1)[
0
]
new_name = original_name_without_extension + "_output.wav"
output_path = os.path.join(os.path.dirname(input_audio_path), new_name)
return output_path
def change_choices():
names = [
os.path.join(root, file)
for root, _, files in os.walk(model_root_relative, topdown=False)
for file in files
if (
file.endswith((".pth", ".onnx"))
and not (file.startswith("G_") or file.startswith("D_"))
)
]
indexes_list = [
os.path.join(root, name)
for root, _, files in os.walk(model_root_relative, topdown=False)
for name in files
if name.endswith(".index") and "trained" not in name
]
audio_paths = [
os.path.join(root, name)
for root, _, files in os.walk(audio_root_relative, topdown=False)
for name in files
if name.endswith(tuple(sup_audioext))
and root == audio_root_relative
and "_output" not in name
]
return (
{"choices": sorted(names), "__type__": "update"},
{"choices": sorted(indexes_list), "__type__": "update"},
{"choices": sorted(audio_paths), "__type__": "update"},
)
def get_indexes():
indexes_list = [
os.path.join(dirpath, filename)
for dirpath, _, filenames in os.walk(model_root_relative)
for filename in filenames
if filename.endswith(".index") and "trained" not in filename
]
return indexes_list if indexes_list else ""
def save_to_wav(record_button):
if record_button is None:
pass
else:
path_to_file = record_button
new_name = datetime.datetime.now().strftime("%Y-%m-%d_%H-%M-%S") + ".wav"
target_path = os.path.join(audio_root_relative, os.path.basename(new_name))
shutil.move(path_to_file, target_path)
return target_path, output_path_fn(target_path)
def save_to_wav2(upload_audio):
file_path = upload_audio
formated_name = format_title(os.path.basename(file_path))
target_path = os.path.join(audio_root_relative, formated_name)
if os.path.exists(target_path):
os.remove(target_path)
shutil.copy(file_path, target_path)
return target_path, output_path_fn(target_path)
def delete_outputs():
gr.Info(f"Outputs cleared!")
for root, _, files in os.walk(audio_root_relative, topdown=False):
for name in files:
if name.endswith(tuple(sup_audioext)) and name.__contains__("_output"):
os.remove(os.path.join(root, name))
def match_index(model_file_value):
if model_file_value:
model_folder = os.path.dirname(model_file_value)
index_files = get_indexes()
for index_file in index_files:
if os.path.dirname(index_file) == model_folder:
return index_file
return ""
# Inference tab
def inference_tab():
default_weight = random.choice(names) if names else None
with gr.Row():
with gr.Row():
model_file = gr.Dropdown(
label=i18n("Voice Model"),
info=i18n("Select the voice model to use for the conversion."),
choices=sorted(names, key=lambda path: os.path.getsize(path)),
interactive=True,
value=default_weight,
allow_custom_value=True,
)
index_file = gr.Dropdown(
label=i18n("Index File"),
info=i18n("Select the index file to use for the conversion."),
choices=get_indexes(),
value=match_index(default_weight) if default_weight else "",
interactive=True,
allow_custom_value=True,
)
with gr.Column():
refresh_button = gr.Button(i18n("Refresh"))
unload_button = gr.Button(i18n("Unload Voice"))
unload_button.click(
fn=lambda: (
{"value": "", "__type__": "update"},
{"value": "", "__type__": "update"},
),
inputs=[],
outputs=[model_file, index_file],
)
model_file.select(
fn=lambda model_file_value: match_index(model_file_value),
inputs=[model_file],
outputs=[index_file],
)
# Single inference tab
with gr.Tab(i18n("Single")):
with gr.Column():
upload_audio = gr.Audio(
label=i18n("Upload Audio"), type="filepath", editable=False
)
with gr.Row():
audio = gr.Dropdown(
label=i18n("Select Audio"),
info=i18n("Select the audio to convert."),
choices=sorted(audio_paths),
value=audio_paths[0] if audio_paths else "",
interactive=True,
allow_custom_value=True,
)
with gr.Accordion(i18n("Advanced Settings"), open=False):
with gr.Column():
clear_outputs_infer = gr.Button(
i18n("Clear Outputs (Deletes all audios in assets/audios)")
)
output_path = gr.Textbox(
label=i18n("Output Path"),
placeholder=i18n("Enter output path"),
info=i18n(
"The path where the output audio will be saved, by default in assets/audios/output.wav"
),
value=(
output_path_fn(audio_paths[0])
if audio_paths
else os.path.join(now_dir, "assets", "audios", "output.wav")
),
interactive=True,
)
export_format = gr.Radio(
label=i18n("Export Format"),
info=i18n("Select the format to export the audio."),
choices=["WAV", "MP3", "FLAC", "OGG", "M4A"],
value="WAV",
interactive=True,
)
split_audio = gr.Checkbox(
label=i18n("Split Audio"),
info=i18n(
"Split the audio into chunks for inference to obtain better results in some cases."
),
visible=True,
value=False,
interactive=True,
)
autotune = gr.Checkbox(
label=i18n("Autotune"),
info=i18n(
"Apply a soft autotune to your inferences, recommended for singing conversions."
),
visible=True,
value=False,
interactive=True,
)
clean_audio = gr.Checkbox(
label=i18n("Clean Audio"),
info=i18n(
"Clean your audio output using noise detection algorithms, recommended for speaking audios."
),
visible=True,
value=False,
interactive=True,
)
clean_strength = gr.Slider(
minimum=0,
maximum=1,
label=i18n("Clean Strength"),
info=i18n(
"Set the clean-up level to the audio you want, the more you increase it the more it will clean up, but it is possible that the audio will be more compressed."
),
visible=False,
value=0.5,
interactive=True,
)
pitch = gr.Slider(
minimum=-24,
maximum=24,
step=1,
label=i18n("Pitch"),
info=i18n(
"Set the pitch of the audio, the higher the value, the higher the pitch."
),
value=0,
interactive=True,
)
filter_radius = gr.Slider(
minimum=0,
maximum=7,
label=i18n("Filter Radius"),
info=i18n(
"If the number is greater than or equal to three, employing median filtering on the collected tone results has the potential to decrease respiration."
),
value=3,
step=1,
interactive=True,
)
index_rate = gr.Slider(
minimum=0,
maximum=1,
label=i18n("Search Feature Ratio"),
info=i18n(
"Influence exerted by the index file; a higher value corresponds to greater influence. However, opting for lower values can help mitigate artifacts present in the audio."
),
value=0.75,
interactive=True,
)
rms_mix_rate = gr.Slider(
minimum=0,
maximum=1,
label=i18n("Volume Envelope"),
info=i18n(
"Substitute or blend with the volume envelope of the output. The closer the ratio is to 1, the more the output envelope is employed."
),
value=1,
interactive=True,
)
protect = gr.Slider(
minimum=0,
maximum=0.5,
label=i18n("Protect Voiceless Consonants"),
info=i18n(
"Safeguard distinct consonants and breathing sounds to prevent electro-acoustic tearing and other artifacts. Pulling the parameter to its maximum value of 0.5 offers comprehensive protection. However, reducing this value might decrease the extent of protection while potentially mitigating the indexing effect."
),
value=0.5,
interactive=True,
)
hop_length = gr.Slider(
minimum=1,
maximum=512,
step=1,
label=i18n("Hop Length"),
info=i18n(
"Denotes the duration it takes for the system to transition to a significant pitch change. Smaller hop lengths require more time for inference but tend to yield higher pitch accuracy."
),
visible=False,
value=128,
interactive=True,
)
with gr.Column():
f0method = gr.Radio(
label=i18n("Pitch extraction algorithm"),
info=i18n(
"Pitch extraction algorithm to use for the audio conversion. The default algorithm is rmvpe, which is recommended for most cases."
),
choices=[
"pm",
"harvest",
"dio",
"crepe",
"crepe-tiny",
"rmvpe",
"fcpe",
"hybrid[rmvpe+fcpe]",
],
value="rmvpe",
interactive=True,
)
convert_button1 = gr.Button(i18n("Convert"))
with gr.Row(): # Defines output info + output audio download after conversion
vc_output1 = gr.Textbox(
label=i18n("Output Information"),
info=i18n("The output information will be displayed here."),
)
vc_output2 = gr.Audio(label=i18n("Export Audio"))
# Batch inference tab
with gr.Tab(i18n("Batch")):
with gr.Row():
with gr.Column():
input_folder_batch = gr.Textbox(
label=i18n("Input Folder"),
info=i18n("Select the folder containing the audios to convert."),
placeholder=i18n("Enter input path"),
value=os.path.join(now_dir, "assets", "audios"),
interactive=True,
)
output_folder_batch = gr.Textbox(
label=i18n("Output Folder"),
info=i18n(
"Select the folder where the output audios will be saved."
),
placeholder=i18n("Enter output path"),
value=os.path.join(now_dir, "assets", "audios"),
interactive=True,
)
with gr.Accordion(i18n("Advanced Settings"), open=False):
with gr.Column():
clear_outputs_batch = gr.Button(
i18n("Clear Outputs (Deletes all audios in assets/audios)")
)
export_format_batch = gr.Radio(
label=i18n("Export Format"),
info=i18n("Select the format to export the audio."),
choices=["WAV", "MP3", "FLAC", "OGG", "M4A"],
value="WAV",
interactive=True,
)
split_audio_batch = gr.Checkbox(
label=i18n("Split Audio"),
info=i18n(
"Split the audio into chunks for inference to obtain better results in some cases."
),
visible=True,
value=False,
interactive=True,
)
autotune_batch = gr.Checkbox(
label=i18n("Autotune"),
info=i18n(
"Apply a soft autotune to your inferences, recommended for singing conversions."
),
visible=True,
value=False,
interactive=True,
)
clean_audio_batch = gr.Checkbox(
label=i18n("Clean Audio"),
info=i18n(
"Clean your audio output using noise detection algorithms, recommended for speaking audios."
),
visible=True,
value=False,
interactive=True,
)
clean_strength_batch = gr.Slider(
minimum=0,
maximum=1,
label=i18n("Clean Strength"),
info=i18n(
"Set the clean-up level to the audio you want, the more you increase it the more it will clean up, but it is possible that the audio will be more compressed."
),
visible=False,
value=0.5,
interactive=True,
)
pitch_batch = gr.Slider(
minimum=-24,
maximum=24,
step=1,
label=i18n("Pitch"),
info=i18n(
"Set the pitch of the audio, the higher the value, the higher the pitch."
),
value=0,
interactive=True,
)
filter_radius_batch = gr.Slider(
minimum=0,
maximum=7,
label=i18n("Filter Radius"),
info=i18n(
"If the number is greater than or equal to three, employing median filtering on the collected tone results has the potential to decrease respiration."
),
value=3,
step=1,
interactive=True,
)
index_rate_batch = gr.Slider(
minimum=0,
maximum=1,
label=i18n("Search Feature Ratio"),
info=i18n(
"Influence exerted by the index file; a higher value corresponds to greater influence. However, opting for lower values can help mitigate artifacts present in the audio."
),
value=0.75,
interactive=True,
)
rms_mix_rate_batch = gr.Slider(
minimum=0,
maximum=1,
label=i18n("Volume Envelope"),
info=i18n(
"Substitute or blend with the volume envelope of the output. The closer the ratio is to 1, the more the output envelope is employed."
),
value=1,
interactive=True,
)
protect_batch = gr.Slider(
minimum=0,
maximum=0.5,
label=i18n("Protect Voiceless Consonants"),
info=i18n(
"Safeguard distinct consonants and breathing sounds to prevent electro-acoustic tearing and other artifacts. Pulling the parameter to its maximum value of 0.5 offers comprehensive protection. However, reducing this value might decrease the extent of protection while potentially mitigating the indexing effect."
),
value=0.5,
interactive=True,
)
hop_length_batch = gr.Slider(
minimum=1,
maximum=512,
step=1,
label=i18n("Hop Length"),
info=i18n(
"Denotes the duration it takes for the system to transition to a significant pitch change. Smaller hop lengths require more time for inference but tend to yield higher pitch accuracy."
),
visible=False,
value=128,
interactive=True,
)
with gr.Column():
f0method_batch = gr.Radio(
label=i18n("Pitch extraction algorithm"),
info=i18n(
"Pitch extraction algorithm to use for the audio conversion. The default algorithm is rmvpe, which is recommended for most cases."
),
choices=[
"pm",
"harvest",
"dio",
"crepe",
"crepe-tiny",
"rmvpe",
"fcpe",
"hybrid[rmvpe+fcpe]",
],
value="rmvpe",
interactive=True,
)
convert_button2 = gr.Button(i18n("Convert"))
with gr.Row(): # Defines output info + output audio download after conversion
vc_output3 = gr.Textbox(
label=i18n("Output Information"),
info=i18n("The output information will be displayed here."),
)
def toggle_visible(checkbox):
return {"visible": checkbox, "__type__": "update"}
def toggle_visible_hop_length(f0method):
if f0method == "crepe" or f0method == "crepe-tiny":
return {"visible": True, "__type__": "update"}
return {"visible": False, "__type__": "update"}
clean_audio.change(
fn=toggle_visible,
inputs=[clean_audio],
outputs=[clean_strength],
)
clean_audio_batch.change(
fn=toggle_visible,
inputs=[clean_audio_batch],
outputs=[clean_strength_batch],
)
f0method.change(
fn=toggle_visible_hop_length,
inputs=[f0method],
outputs=[hop_length],
)
f0method_batch.change(
fn=toggle_visible_hop_length,
inputs=[f0method_batch],
outputs=[hop_length_batch],
)
refresh_button.click(
fn=change_choices,
inputs=[],
outputs=[model_file, index_file, audio],
)
audio.change(
fn=output_path_fn,
inputs=[audio],
outputs=[output_path],
)
upload_audio.upload(
fn=save_to_wav2,
inputs=[upload_audio],
outputs=[audio, output_path],
)
upload_audio.stop_recording(
fn=save_to_wav,
inputs=[upload_audio],
outputs=[audio, output_path],
)
clear_outputs_infer.click(
fn=delete_outputs,
inputs=[],
outputs=[],
)
clear_outputs_batch.click(
fn=delete_outputs,
inputs=[],
outputs=[],
)
convert_button1.click(
fn=run_infer_script,
inputs=[
pitch,
filter_radius,
index_rate,
rms_mix_rate,
protect,
hop_length,
f0method,
audio,
output_path,
model_file,
index_file,
split_audio,
autotune,
clean_audio,
clean_strength,
export_format,
],
outputs=[vc_output1, vc_output2],
)
convert_button2.click(
fn=run_batch_infer_script,
inputs=[
pitch_batch,
filter_radius_batch,
index_rate_batch,
rms_mix_rate_batch,
protect_batch,
hop_length_batch,
f0method_batch,
input_folder_batch,
output_folder_batch,
model_file,
index_file,
split_audio_batch,
autotune_batch,
clean_audio_batch,
clean_strength_batch,
export_format_batch,
],
outputs=[vc_output3],
)
|