Spaces:
Build error
Build error
Update app.py
Browse files
app.py
CHANGED
|
@@ -23,16 +23,6 @@ from mutagen.mp3 import MP3
|
|
| 23 |
from mutagen.id3 import ID3, TIT2, TPE1, TALB, TYER
|
| 24 |
from TTS.api import TTS
|
| 25 |
import pickle
|
| 26 |
-
import subprocess
|
| 27 |
-
|
| 28 |
-
# Install OpenVoice from GitHub during startup
|
| 29 |
-
try:
|
| 30 |
-
import openvoice
|
| 31 |
-
except ImportError:
|
| 32 |
-
print("Installing OpenVoice from GitHub...")
|
| 33 |
-
subprocess.run(["pip", "install", "git+https://github.com/myshell-ai/OpenVoice.git"])
|
| 34 |
-
from openvoice.se_extractor import get_se
|
| 35 |
-
from openvoice.api import ToneColorConverter
|
| 36 |
|
| 37 |
# Suppress warnings
|
| 38 |
warnings.filterwarnings("ignore")
|
|
@@ -292,23 +282,6 @@ def generate_tts(text):
|
|
| 292 |
tts.tts_to_file(text=text, file_path=out_path)
|
| 293 |
return out_path
|
| 294 |
|
| 295 |
-
# === Save/Load Project File (.aiproj) ===
|
| 296 |
-
def save_project(audio_path, preset_name, effects):
|
| 297 |
-
project_data = {
|
| 298 |
-
"audio": AudioSegment.from_file(audio_path).raw_data,
|
| 299 |
-
"preset": preset_name,
|
| 300 |
-
"effects": effects
|
| 301 |
-
}
|
| 302 |
-
out_path = os.path.join(tempfile.gettempdir(), "project.aiproj")
|
| 303 |
-
with open(out_path, "wb") as f:
|
| 304 |
-
pickle.dump(project_data, f)
|
| 305 |
-
return out_path
|
| 306 |
-
|
| 307 |
-
def load_project(project_file):
|
| 308 |
-
with open(project_file.name, "rb") as f:
|
| 309 |
-
data = pickle.load(f)
|
| 310 |
-
return data["preset"], data["effects"]
|
| 311 |
-
|
| 312 |
# === Trim Silence Automatically (VAD) ===
|
| 313 |
def detect_silence(audio_file, silence_threshold=-50.0, min_silence_len=1000):
|
| 314 |
audio = AudioSegment.from_file(audio_file)
|
|
@@ -336,27 +309,244 @@ def mix_tracks(track1, track2, volume_offset=0):
|
|
| 336 |
mixed.export(out_path, format="wav")
|
| 337 |
return out_path
|
| 338 |
|
| 339 |
-
# ===
|
| 340 |
-
|
| 341 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 342 |
|
| 343 |
-
|
|
|
|
|
|
|
|
|
|
| 344 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 345 |
def clone_voice(source_audio, target_audio, text):
|
| 346 |
-
|
| 347 |
-
|
| 348 |
-
|
| 349 |
-
|
| 350 |
-
|
| 351 |
-
|
| 352 |
-
|
| 353 |
-
|
| 354 |
-
|
| 355 |
-
|
| 356 |
-
|
| 357 |
-
|
| 358 |
-
|
| 359 |
-
|
| 360 |
-
|
| 361 |
-
|
| 362 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 23 |
from mutagen.id3 import ID3, TIT2, TPE1, TALB, TYER
|
| 24 |
from TTS.api import TTS
|
| 25 |
import pickle
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 26 |
|
| 27 |
# Suppress warnings
|
| 28 |
warnings.filterwarnings("ignore")
|
|
|
|
| 282 |
tts.tts_to_file(text=text, file_path=out_path)
|
| 283 |
return out_path
|
| 284 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 285 |
# === Trim Silence Automatically (VAD) ===
|
| 286 |
def detect_silence(audio_file, silence_threshold=-50.0, min_silence_len=1000):
|
| 287 |
audio = AudioSegment.from_file(audio_file)
|
|
|
|
| 309 |
mixed.export(out_path, format="wav")
|
| 310 |
return out_path
|
| 311 |
|
| 312 |
+
# === Save/Load Project File (.aiproj) ===
|
| 313 |
+
def save_project(audio_path, preset_name, effects):
|
| 314 |
+
project_data = {
|
| 315 |
+
"audio": AudioSegment.from_file(audio_path).raw_data,
|
| 316 |
+
"preset": preset_name,
|
| 317 |
+
"effects": effects
|
| 318 |
+
}
|
| 319 |
+
out_path = os.path.join(tempfile.gettempdir(), "project.aiproj")
|
| 320 |
+
with open(out_path, "wb") as f:
|
| 321 |
+
pickle.dump(project_data, f)
|
| 322 |
+
return out_path
|
| 323 |
|
| 324 |
+
def load_project(project_file):
|
| 325 |
+
with open(project_file.name, "rb") as f:
|
| 326 |
+
data = pickle.load(f)
|
| 327 |
+
return data["preset"], data["effects"]
|
| 328 |
|
| 329 |
+
# === Auto-Save / Resume Sessions ===
|
| 330 |
+
def save_or_resume_session(audio, preset, effects, action="save"):
|
| 331 |
+
if action == "save":
|
| 332 |
+
return {"audio": audio, "preset": preset, "effects": effects}, None, None, None
|
| 333 |
+
elif action == "load" and isinstance(audio, dict):
|
| 334 |
+
return (
|
| 335 |
+
None,
|
| 336 |
+
audio.get("audio"),
|
| 337 |
+
audio.get("preset"),
|
| 338 |
+
audio.get("effects")
|
| 339 |
+
)
|
| 340 |
+
return None, None, None, None
|
| 341 |
+
|
| 342 |
+
# === Voice Cloning β Fallback Version for Hugging Face ===
|
| 343 |
def clone_voice(source_audio, target_audio, text):
|
| 344 |
+
print("β οΈ Voice cloning not available in browser version β use local install for full support")
|
| 345 |
+
return generate_tts(text)
|
| 346 |
+
|
| 347 |
+
# === UI Setup ===
|
| 348 |
+
effect_options = [
|
| 349 |
+
"Noise Reduction",
|
| 350 |
+
"Compress Dynamic Range",
|
| 351 |
+
"Add Reverb",
|
| 352 |
+
"Pitch Shift",
|
| 353 |
+
"Echo",
|
| 354 |
+
"Stereo Widening",
|
| 355 |
+
"Bass Boost",
|
| 356 |
+
"Treble Boost",
|
| 357 |
+
"Normalize"
|
| 358 |
+
]
|
| 359 |
+
|
| 360 |
+
with gr.Blocks(title="AI Audio Studio", css="style.css") as demo:
|
| 361 |
+
gr.Markdown("## π§ Ultimate AI Audio Studio\nUpload, edit, export β powered by AI!")
|
| 362 |
+
|
| 363 |
+
# --- Single File Studio ---
|
| 364 |
+
with gr.Tab("π΅ Single File Studio"):
|
| 365 |
+
gr.Interface(
|
| 366 |
+
fn=process_audio,
|
| 367 |
+
inputs=[
|
| 368 |
+
gr.Audio(label="Upload Audio", type="filepath"),
|
| 369 |
+
gr.CheckboxGroup(choices=effect_options, label="Apply Effects in Order"),
|
| 370 |
+
gr.Checkbox(label="Isolate Vocals After Effects"),
|
| 371 |
+
gr.Dropdown(choices=preset_names, label="Select Preset", value=preset_names[0] if preset_names else None),
|
| 372 |
+
gr.Dropdown(choices=["MP3", "WAV"], label="Export Format", value="MP3")
|
| 373 |
+
],
|
| 374 |
+
outputs=[
|
| 375 |
+
gr.Audio(label="Processed Audio", type="filepath"),
|
| 376 |
+
gr.Image(label="Waveform Preview"),
|
| 377 |
+
gr.Textbox(label="Session Log (JSON)", lines=5),
|
| 378 |
+
gr.Textbox(label="Detected Genre", lines=1),
|
| 379 |
+
gr.Textbox(label="Status", value="β
Ready", lines=1)
|
| 380 |
+
],
|
| 381 |
+
title="Edit One File at a Time",
|
| 382 |
+
description="Apply effects, preview waveform, and get full session log.",
|
| 383 |
+
flagging_mode="never",
|
| 384 |
+
submit_btn="Process Audio",
|
| 385 |
+
clear_btn=None
|
| 386 |
+
)
|
| 387 |
+
|
| 388 |
+
# --- Batch Processing ---
|
| 389 |
+
with gr.Tab("π Batch Processing"):
|
| 390 |
+
gr.Interface(
|
| 391 |
+
fn=batch_process_audio,
|
| 392 |
+
inputs=[
|
| 393 |
+
gr.File(label="Upload Multiple Files", file_count="multiple"),
|
| 394 |
+
gr.CheckboxGroup(choices=effect_options, label="Apply Effects in Order"),
|
| 395 |
+
gr.Checkbox(label="Isolate Vocals After Effects"),
|
| 396 |
+
gr.Dropdown(choices=preset_names, label="Select Preset", value=preset_names[0] if preset_names else None),
|
| 397 |
+
gr.Dropdown(choices=["MP3", "WAV"], label="Export Format", value="MP3")
|
| 398 |
+
],
|
| 399 |
+
outputs=[
|
| 400 |
+
gr.File(label="Download ZIP of All Processed Files"),
|
| 401 |
+
gr.Textbox(label="Status", value="β
Ready", lines=1)
|
| 402 |
+
],
|
| 403 |
+
title="Batch Audio Processor",
|
| 404 |
+
description="Upload multiple files, apply effects in bulk, and download all results in a single ZIP.",
|
| 405 |
+
flagging_mode="never",
|
| 406 |
+
submit_btn="Process All Files",
|
| 407 |
+
clear_btn=None
|
| 408 |
+
)
|
| 409 |
+
|
| 410 |
+
# --- Remix Mode ---
|
| 411 |
+
with gr.Tab("π Remix Mode"):
|
| 412 |
+
gr.Interface(
|
| 413 |
+
fn=stem_split,
|
| 414 |
+
inputs=gr.Audio(label="Upload Music Track", type="filepath"),
|
| 415 |
+
outputs=[
|
| 416 |
+
gr.File(label="Vocals"),
|
| 417 |
+
gr.File(label="Drums"),
|
| 418 |
+
gr.File(label="Bass"),
|
| 419 |
+
gr.File(label="Other")
|
| 420 |
+
],
|
| 421 |
+
title="Split Into Drums, Bass, Vocals, and More",
|
| 422 |
+
description="Use AI to separate musical elements like vocals, drums, and bass.",
|
| 423 |
+
flagging_mode="never",
|
| 424 |
+
clear_btn=None
|
| 425 |
+
)
|
| 426 |
+
|
| 427 |
+
# --- Transcribe & Edit ---
|
| 428 |
+
with gr.Tab("π Transcribe & Edit"):
|
| 429 |
+
gr.Interface(
|
| 430 |
+
fn=transcribe_audio,
|
| 431 |
+
inputs=gr.Audio(label="Upload Audio", type="filepath"),
|
| 432 |
+
outputs=gr.Textbox(label="Transcribed Text", lines=10),
|
| 433 |
+
title="Transcribe Spoken Content",
|
| 434 |
+
description="Convert voice to text and edit it before exporting again."
|
| 435 |
+
)
|
| 436 |
+
|
| 437 |
+
# --- TTS Voice Generator ---
|
| 438 |
+
with gr.Tab("π¬ TTS Voice Generator"):
|
| 439 |
+
gr.Interface(
|
| 440 |
+
fn=generate_tts,
|
| 441 |
+
inputs=gr.Textbox(label="Enter Text", lines=5),
|
| 442 |
+
outputs=gr.Audio(label="Generated Speech", type="filepath"),
|
| 443 |
+
title="Text-to-Speech Generator",
|
| 444 |
+
description="Type anything and turn it into natural-sounding speech."
|
| 445 |
+
)
|
| 446 |
+
|
| 447 |
+
# --- VAD β Detect & Remove Silence ===
|
| 448 |
+
with gr.Tab("βοΈ Trim Silence Automatically"):
|
| 449 |
+
gr.Interface(
|
| 450 |
+
fn=detect_silence,
|
| 451 |
+
inputs=[
|
| 452 |
+
gr.File(label="Upload Track"),
|
| 453 |
+
gr.Slider(minimum=-100, maximum=-10, value=-50, label="Silence Threshold (dB)"),
|
| 454 |
+
gr.Number(label="Min Silence Length (ms)", value=1000)
|
| 455 |
+
],
|
| 456 |
+
outputs=gr.File(label="Trimmed Output"),
|
| 457 |
+
title="Auto-Detect & Remove Silence",
|
| 458 |
+
description="Detect and trim silence at start/end or between words"
|
| 459 |
+
)
|
| 460 |
+
|
| 461 |
+
# --- Load/Save Project File (.aiproj) ===
|
| 462 |
+
with gr.Tab("π Save/Load Project"):
|
| 463 |
+
gr.Interface(
|
| 464 |
+
fn=save_project,
|
| 465 |
+
inputs=[
|
| 466 |
+
gr.File(label="Original Audio"),
|
| 467 |
+
gr.Dropdown(choices=preset_names, label="Used Preset", value=preset_names[0]),
|
| 468 |
+
gr.CheckboxGroup(choices=effect_options, label="Applied Effects")
|
| 469 |
+
],
|
| 470 |
+
outputs=gr.File(label="Project File (.aiproj)"),
|
| 471 |
+
title="Save Everything Together",
|
| 472 |
+
description="Save your session, effects, and settings in one file to reuse later."
|
| 473 |
+
)
|
| 474 |
+
|
| 475 |
+
gr.Interface(
|
| 476 |
+
fn=load_project,
|
| 477 |
+
inputs=gr.File(label="Upload .aiproj File"),
|
| 478 |
+
outputs=[
|
| 479 |
+
gr.Dropdown(choices=preset_names, label="Loaded Preset"),
|
| 480 |
+
gr.CheckboxGroup(choices=effect_options, label="Loaded Effects")
|
| 481 |
+
],
|
| 482 |
+
title="Resume Last Project",
|
| 483 |
+
description="Load your saved session"
|
| 484 |
+
)
|
| 485 |
+
|
| 486 |
+
# --- Auto-Save / Resume Sessions ===
|
| 487 |
+
session_state = gr.State()
|
| 488 |
+
|
| 489 |
+
with gr.Tab("π§Ύ Auto-Save & Resume"):
|
| 490 |
+
gr.Markdown("Save your current state and resume editing later.")
|
| 491 |
+
|
| 492 |
+
action_radio = gr.Radio(["save", "load"], label="Action", value="save")
|
| 493 |
+
audio_input = gr.Audio(label="Upload or Load Audio", type="filepath")
|
| 494 |
+
preset_dropdown = gr.Dropdown(choices=preset_names, label="Used Preset", value=preset_names[0] if preset_names else None)
|
| 495 |
+
effect_checkbox = gr.CheckboxGroup(choices=effect_options, label="Applied Effects")
|
| 496 |
+
save_btn = gr.Button("Save or Load Session")
|
| 497 |
+
|
| 498 |
+
loaded_audio = gr.Audio(label="Loaded Audio", type="filepath")
|
| 499 |
+
loaded_preset = gr.Dropdown(choices=preset_names, label="Loaded Preset")
|
| 500 |
+
loaded_effects = gr.CheckboxGroup(choices=effect_options, label="Loaded Effects")
|
| 501 |
+
|
| 502 |
+
save_btn.click(
|
| 503 |
+
fn=save_or_resume_session,
|
| 504 |
+
inputs=[audio_input, preset_dropdown, effect_checkbox, action_radio],
|
| 505 |
+
outputs=[session_state, loaded_audio, loaded_preset, loaded_effects]
|
| 506 |
+
)
|
| 507 |
+
|
| 508 |
+
# --- Mix Two Tracks ===
|
| 509 |
+
with gr.Tab("π Mix Two Tracks"):
|
| 510 |
+
gr.Interface(
|
| 511 |
+
fn=mix_tracks,
|
| 512 |
+
inputs=[
|
| 513 |
+
gr.File(label="Main Track"),
|
| 514 |
+
gr.File(label="Background Track"),
|
| 515 |
+
gr.Slider(minimum=-10, maximum=10, value=0, label="Volume Offset (dB)")
|
| 516 |
+
],
|
| 517 |
+
outputs=gr.File(label="Mixed Output"),
|
| 518 |
+
title="Overlay Two Tracks",
|
| 519 |
+
description="Mix, blend, or subtract two audio files."
|
| 520 |
+
)
|
| 521 |
+
|
| 522 |
+
# === Voice Style Transfer (Dummy) ===
|
| 523 |
+
def apply_style_transfer(audio_path, mood="Happy"):
|
| 524 |
+
return audio_path
|
| 525 |
+
|
| 526 |
+
with gr.Tab("π§ Voice Style Transfer"):
|
| 527 |
+
gr.Interface(
|
| 528 |
+
fn=apply_style_transfer,
|
| 529 |
+
inputs=[
|
| 530 |
+
gr.Audio(label="Upload Voice Clip", type="filepath"),
|
| 531 |
+
gr.Radio(["Happy", "Sad", "Angry", "Calm"], label="Choose Tone")
|
| 532 |
+
],
|
| 533 |
+
outputs=gr.Audio(label="Stylized Output", type="filepath"),
|
| 534 |
+
title="Change Emotional Tone of Voice",
|
| 535 |
+
description="Shift the emotional style of any voice clip."
|
| 536 |
+
)
|
| 537 |
+
|
| 538 |
+
# --- Voice Cloning (Fallback) ===
|
| 539 |
+
with gr.Tab("π Voice Cloning (Demo)"):
|
| 540 |
+
gr.Interface(
|
| 541 |
+
fn=clone_voice,
|
| 542 |
+
inputs=[
|
| 543 |
+
gr.File(label="Source Voice Clip"),
|
| 544 |
+
gr.File(label="Target Voice Clip"),
|
| 545 |
+
gr.Textbox(label="Text to Clone", lines=5)
|
| 546 |
+
],
|
| 547 |
+
outputs=gr.Audio(label="Cloned Output", type="filepath"),
|
| 548 |
+
title="Replace One Voice With Another (Demo)",
|
| 549 |
+
description="Clone voice from source to target speaker using AI"
|
| 550 |
+
)
|
| 551 |
+
|
| 552 |
+
demo.launch()
|