Spaces:
Runtime error
Runtime error
File size: 15,269 Bytes
a8c39f5 |
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 |
import json
import os
import random
import sys
import gradio as gr
now_dir = os.getcwd()
sys.path.append(now_dir)
from assets.i18n.i18n import I18nAuto
from core import run_tts_script
from tabs.inference.inference import (
change_choices,
create_folder_and_move_files,
get_indexes,
get_speakers_id,
match_index,
names,
refresh_embedders_folders,
)
i18n = I18nAuto()
default_weight = random.choice(names) if names else ""
with open(
os.path.join("rvc", "lib", "tools", "tts_voices.json"), "r", encoding="utf-8"
) as file:
tts_voices_data = json.load(file)
short_names = [voice.get("ShortName", "") for voice in tts_voices_data]
def process_input(file_path):
try:
with open(file_path, "r", encoding="utf-8") as file:
file.read()
gr.Info(f"The file has been loaded!")
return file_path, file_path
except UnicodeDecodeError:
gr.Info(f"The file has to be in UTF-8 encoding.")
return None, None
# TTS tab
def tts_tab():
with gr.Column():
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,
)
best_default_index_path = match_index(model_file.value)
index_file = gr.Dropdown(
label=i18n("Index File"),
info=i18n("Select the index file to use for the conversion."),
choices=get_indexes(),
value=best_default_index_path,
interactive=True,
allow_custom_value=True,
)
with gr.Row():
unload_button = gr.Button(i18n("Unload Voice"))
refresh_button = gr.Button(i18n("Refresh"))
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],
)
gr.Markdown(
i18n(
f"Applio is a Speech-to-Speech conversion software, utilizing EdgeTTS as middleware for running the Text-to-Speech (TTS) component. Read more about it [here!](https://docs.applio.org/getting-started/tts#disclaimer)"
)
)
tts_voice = gr.Dropdown(
label=i18n("TTS Voices"),
info=i18n("Select the TTS voice to use for the conversion."),
choices=short_names,
interactive=True,
value=None,
)
tts_rate = gr.Slider(
minimum=-100,
maximum=100,
step=1,
label=i18n("TTS Speed"),
info=i18n("Increase or decrease TTS speed."),
value=0,
interactive=True,
)
with gr.Tabs():
with gr.Tab(label="Text to Speech"):
tts_text = gr.Textbox(
label=i18n("Text to Synthesize"),
info=i18n("Enter the text to synthesize."),
placeholder=i18n("Enter text to synthesize"),
lines=3,
)
with gr.Tab(label="File to Speech"):
txt_file = gr.File(
label=i18n("Upload a .txt file"),
type="filepath",
)
input_tts_path = gr.Textbox(
label=i18n("Input path for text file"),
placeholder=i18n(
"The path to the text file that contains content for text to speech."
),
value="",
interactive=True,
)
with gr.Accordion(i18n("Advanced Settings"), open=False):
with gr.Column():
output_tts_path = gr.Textbox(
label=i18n("Output Path for TTS Audio"),
placeholder=i18n("Enter output path"),
value=os.path.join(now_dir, "assets", "audios", "tts_output.wav"),
interactive=True,
)
output_rvc_path = gr.Textbox(
label=i18n("Output Path for RVC Audio"),
placeholder=i18n("Enter output path"),
value=os.path.join(now_dir, "assets", "audios", "tts_rvc_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,
)
sid = gr.Dropdown(
label=i18n("Speaker ID"),
info=i18n("Select the speaker ID to use for the conversion."),
choices=get_speakers_id(model_file.value),
value=0,
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,
)
autotune_strength = gr.Slider(
minimum=0,
maximum=1,
label=i18n("Autotune Strength"),
info=i18n(
"Set the autotune strength - the more you increase it the more it will snap to the chromatic grid."
),
visible=False,
value=1,
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=True,
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=True,
value=0.5,
interactive=True,
)
upscale_audio = gr.Checkbox(
label=i18n("Upscale Audio"),
info=i18n(
"Upscale the audio to a higher quality, recommended for low-quality audios. (It could take longer to process the audio)"
),
visible=True,
value=False,
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."
),
value=128,
interactive=True,
)
f0_method = 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=[
"crepe",
"crepe-tiny",
"rmvpe",
"fcpe",
"hybrid[rmvpe+fcpe]",
],
value="rmvpe",
interactive=True,
)
embedder_model = gr.Radio(
label=i18n("Embedder Model"),
info=i18n("Model used for learning speaker embedding."),
choices=[
"contentvec",
"chinese-hubert-base",
"japanese-hubert-base",
"korean-hubert-base",
"custom",
],
value="contentvec",
interactive=True,
)
with gr.Column(visible=False) as embedder_custom:
with gr.Accordion(i18n("Custom Embedder"), open=True):
with gr.Row():
embedder_model_custom = gr.Dropdown(
label=i18n("Select Custom Embedder"),
choices=refresh_embedders_folders(),
interactive=True,
allow_custom_value=True,
)
refresh_embedders_button = gr.Button(i18n("Refresh embedders"))
folder_name_input = gr.Textbox(
label=i18n("Folder Name"), interactive=True
)
with gr.Row():
bin_file_upload = gr.File(
label=i18n("Upload .bin"),
type="filepath",
interactive=True,
)
config_file_upload = gr.File(
label=i18n("Upload .json"),
type="filepath",
interactive=True,
)
move_files_button = gr.Button(
i18n("Move files to custom embedder folder")
)
f0_file = gr.File(
label=i18n(
"The f0 curve represents the variations in the base frequency of a voice over time, showing how pitch rises and falls."
),
visible=True,
)
convert_button = gr.Button(i18n("Convert"))
with gr.Row():
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"))
def toggle_visible(checkbox):
return {"visible": checkbox, "__type__": "update"}
def toggle_visible_embedder_custom(embedder_model):
if embedder_model == "custom":
return {"visible": True, "__type__": "update"}
return {"visible": False, "__type__": "update"}
autotune.change(
fn=toggle_visible,
inputs=[autotune],
outputs=[autotune_strength],
)
clean_audio.change(
fn=toggle_visible,
inputs=[clean_audio],
outputs=[clean_strength],
)
refresh_button.click(
fn=change_choices,
inputs=[model_file],
outputs=[model_file, index_file, sid],
)
txt_file.upload(
fn=process_input,
inputs=[txt_file],
outputs=[input_tts_path, txt_file],
)
embedder_model.change(
fn=toggle_visible_embedder_custom,
inputs=[embedder_model],
outputs=[embedder_custom],
)
move_files_button.click(
fn=create_folder_and_move_files,
inputs=[folder_name_input, bin_file_upload, config_file_upload],
outputs=[],
)
refresh_embedders_button.click(
fn=lambda: gr.update(choices=refresh_embedders_folders()),
inputs=[],
outputs=[embedder_model_custom],
)
convert_button.click(
fn=run_tts_script,
inputs=[
input_tts_path,
tts_text,
tts_voice,
tts_rate,
pitch,
filter_radius,
index_rate,
rms_mix_rate,
protect,
hop_length,
f0_method,
output_tts_path,
output_rvc_path,
model_file,
index_file,
split_audio,
autotune,
autotune_strength,
clean_audio,
clean_strength,
export_format,
upscale_audio,
f0_file,
embedder_model,
embedder_model_custom,
sid,
],
outputs=[vc_output1, vc_output2],
)
|