File size: 11,879 Bytes
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 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 1a7d583 85d3b29 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 |
import os, sys
import gradio as gr
import regex as re
import json
import random
from core import (
run_tts_script,
)
from assets.i18n.i18n import I18nAuto
i18n = I18nAuto()
now_dir = os.getcwd()
sys.path.append(now_dir)
model_root = os.path.join(now_dir, "logs")
model_root_relative = os.path.relpath(model_root, now_dir)
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
]
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
]
return (
{"choices": sorted(names), "__type__": "update"},
{"choices": sorted(indexes_list), "__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 process_input(file_path):
with open(file_path, "r") as file:
file_contents = file.read()
gr.Info(f"The text from the txt file has been loaded!")
return file_contents, None
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 ""
def tts_tab():
default_weight = random.choice(names) if names else ""
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,
)
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.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],
)
json_path = os.path.join("rvc", "lib", "tools", "tts_voices.json")
with open(json_path, "r") as file:
tts_voices_data = json.load(file)
short_names = [voice.get("ShortName", "") for voice in tts_voices_data]
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_text = gr.Textbox(
label=i18n("Text to Synthesize"),
info=i18n("Enter the text to synthesize."),
placeholder=i18n("Enter text to synthesize"),
lines=3,
)
txt_file = gr.File(
label=i18n("Or you can upload a .txt file"),
type="filepath",
)
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,
)
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=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,
)
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,
)
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"))
def toggle_visible(checkbox):
return {"visible": checkbox, "__type__": "update"}
clean_audio.change(
fn=toggle_visible,
inputs=[clean_audio],
outputs=[clean_strength],
)
refresh_button.click(
fn=change_choices,
inputs=[],
outputs=[model_file, index_file],
)
txt_file.upload(
fn=process_input,
inputs=[txt_file],
outputs=[tts_text, txt_file],
)
convert_button1.click(
fn=run_tts_script,
inputs=[
tts_text,
tts_voice,
pitch,
filter_radius,
index_rate,
rms_mix_rate,
protect,
hop_length,
f0method,
output_tts_path,
output_rvc_path,
model_file,
index_file,
split_audio,
autotune,
clean_audio,
clean_strength,
export_format,
],
outputs=[vc_output1, vc_output2],
)
|