Upload almiq.pth

README.md

@@ -4,7 +4,7 @@ emoji: ⚡
 colorFrom: gray
 colorTo: indigo
 sdk: gradio
-sdk_version: 5.43.1
+sdk_version: 4.28.3
 app_file: app.py
 license: mit
 pinned: true

app.py

@@ -1,5 +1,4 @@
 import os
-import subprocess
 import gradio as gr
 import spaces
 from infer_rvc_python import BaseLoader
@@ -7,7 +6,7 @@ import random
 import logging
 import time
 import soundfile as sf
-from infer_rvc_python.main import download_manager, load_hu_bert, Config
+from infer_rvc_python.main import download_manager
 import zipfile
 import edge_tts
 import asyncio
@@ -20,67 +19,16 @@ from pydub import AudioSegment
 import noisereduce as nr
 import numpy as np
 import urllib.request
-import urllib.parse
-import urllib.error
 import shutil
 import threading
-import argparse
-import sys
-import torch
-import fairseq
-from picklescan.scanner import scan_file_path
-import hashlib
-
-
-ALLOWED_DOMAINS = {"huggingface.co", "hf.co"}
-MODEL_CACHE = {}
-
-parser = argparse.ArgumentParser(description="Run the app with optional sharing")
-parser.add_argument(
-    '--share',
-    action='store_true',
-    help='Enable sharing mode'
-)
-parser.add_argument(
-    '--theme',
-    type=str,
-    default="aliabid94/new-theme",
-    help='Set the theme (default: aliabid94/new-theme)'
-)
-args = parser.parse_args()
-
-IS_COLAB = True if ('google.colab' in sys.modules or args.share) else False
-IS_ZERO_GPU = os.getenv("SPACES_ZERO_GPU")
 
 logging.getLogger("infer_rvc_python").setLevel(logging.ERROR)
 
-torch.serialization.add_safe_globals([fairseq.data.dictionary.Dictionary])
 converter = BaseLoader(only_cpu=False, hubert_path=None, rmvpe_path=None)
-converter.hu_bert_model = load_hu_bert(Config(only_cpu=False), converter.hubert_path)
-
-test_model = "https://huggingface.co/sail-rvc/Aldeano_Minecraft__RVC_V2_-_500_Epochs_/resolve/main/model.pth?download=true, https://huggingface.co/sail-rvc/Aldeano_Minecraft__RVC_V2_-_500_Epochs_/resolve/main/model.index?download=true"
-test_names = ["model.pth", "model.index"]
-
-for url, filename in zip(test_model.split(", "), test_names):
-    try:
-        download_manager(
-            url=url,
-            path=".",
-            extension="",
-            overwrite=False,
-            progress=True,
-        )
-        if not os.path.isfile(filename):
-            raise FileNotFoundError
-    except Exception:
-        with open(filename, "wb") as f:
-            pass
 
 title = "<center><strong><font size='7'>RVC⚡ZERO</font></strong></center>"
-description = "This demo is provided for educational and research purposes only. The authors and contributors of this project do not endorse or encourage any misuse or unethical use of this software. Any use of this software for purposes other than those intended is solely at the user's own risk. The authors and contributors shall not be held responsible for any damages or liabilities arising from the use of this demo inappropriately." if IS_ZERO_GPU else ""
-RESOURCES = "- You can also try `RVC⚡ZERO` in Colab’s free tier, which provides free GPU [link](https://github.com/R3gm/rvc_zero_ui?tab=readme-ov-file#rvczero)."
-theme = args.theme
-delete_cache_time = (3200, 3200) if IS_ZERO_GPU else (86400, 86400)
+description = "This demo is provided for educational and research purposes only. The authors and contributors of this project do not endorse or encourage any misuse or unethical use of this software. Any use of this software for purposes other than those intended is solely at the user's own risk. The authors and contributors shall not be held responsible for any damages or liabilities arising from the use of this demo inappropriately."
+theme = "aliabid94/new-theme"
 
 PITCH_ALGO_OPT = [
     "pm",
@@ -91,54 +39,6 @@ PITCH_ALGO_OPT = [
 ]
 
 
-def check_model_safety(file_path: str):
-    """Statically checks model or archive integrity. Raises ValueError if unsafe."""
-    if not file_path or not os.path.exists(file_path):
-        print(f"Skip file: {file_path}")
-        return
-
-    ext = os.path.splitext(file_path)[1].lower()
-    if ext in (".pth", ".pt", ".bin", ".zip", ".pkl"):
-        try:
-            result = scan_file_path(file_path)
-            if result and result.infected_files > 0:
-                raise ValueError(
-                    f"Integrity check failed: '{os.path.basename(file_path)}' contains unsupported or unsafe structures."
-                )
-        except ValueError:
-            raise
-        except Exception as e:
-            print(f"Integrity check skipped for {file_path}: {e}")
-
-
-def get_file_hash(file_path: str) -> str:
-    hasher = hashlib.sha256()
-    with open(file_path, "rb") as f:
-        while chunk := f.read(65536):
-            hasher.update(chunk)
-    return hasher.hexdigest()
-
-
-async def get_voices_list(proxy=None):
-    """Print all available voices."""
-    from edge_tts import list_voices
-    voices = await list_voices(proxy=proxy)
-    voices = sorted(voices, key=lambda voice: voice["ShortName"])
-
-    table = [
-        {
-            "ShortName": voice["ShortName"],
-            "Gender": voice["Gender"],
-            "ContentCategories": ", ".join(voice["VoiceTag"]["ContentCategories"]),
-            "VoicePersonalities": ", ".join(voice["VoiceTag"]["VoicePersonalities"]),
-            "FriendlyName": voice["FriendlyName"],
-        }
-        for voice in voices
-    ]
-
-    return table
-
-
 def find_files(directory):
     file_paths = []
     for filename in os.listdir(directory):
@@ -162,30 +62,8 @@ def unzip_in_folder(my_zip, my_dir):
 def find_my_model(a_, b_):
 
     if a_ is None or a_.endswith(".pth"):
-        if a_ and a_.endswith(".pth"):
-            check_model_safety(a_)
         return a_, b_
 
-    input_hash = None
-    if a_ and os.path.exists(a_):
-        input_hash = get_file_hash(a_)
-        if b_ and os.path.exists(b_):
-            input_hash += "_" + get_file_hash(b_)
-
-    if input_hash and input_hash in MODEL_CACHE:
-        cached_model, cached_index = MODEL_CACHE[input_hash]
-        model_exists = cached_model and os.path.exists(cached_model)
-        index_exists = (cached_index is None) or os.path.exists(cached_index)
-
-        if model_exists and index_exists:
-            check_model_safety(cached_model)
-            gr.Info(f"Model found: {cached_model}")
-            if cached_index:
-                gr.Info(f"Index found: {cached_index}")
-            return cached_model, cached_index
-        else:
-            del MODEL_CACHE[input_hash]
-
     txt_files = []
     for base_file in [a_, b_]:
         if base_file is not None and base_file.endswith(".txt"):
@@ -197,18 +75,14 @@ def find_my_model(a_, b_):
         with open(txt, 'r') as file:
             first_line = file.readline()
 
-        url_to_download = first_line.strip()
-        ensure_valid_file(url_to_download)
-
         download_manager(
-            url=url_to_download,
+            url=first_line.strip(),
             path=directory,
             extension="",
         )
 
     for f in find_files(directory):
         if f.endswith(".zip"):
-            check_model_safety(f)
             unzip_in_folder(f, directory)
 
     model = None
@@ -217,11 +91,10 @@ def find_my_model(a_, b_):
 
     for ff in end_files:
         if ff.endswith(".pth"):
-            check_model_safety(ff)
-            model = ff
+            model = os.path.join(directory, ff)
             gr.Info(f"Model found: {ff}")
         if ff.endswith(".index"):
-            index = ff
+            index = os.path.join(directory, ff)
             gr.Info(f"Index found: {ff}")
 
     if not model:
@@ -230,115 +103,40 @@ def find_my_model(a_, b_):
     if not index:
         gr.Warning("Index not found")
 
-    if model and input_hash:
-        MODEL_CACHE[input_hash] = (model, index)
-
     return model, index
 
 
-def validate_url(url: str) -> str:
-    """Validate URL protocol and host."""
-    url = url.strip()
-    if not url:
-        raise ValueError("URL cannot be empty.")
+def get_file_size(url):
 
-    parsed = urllib.parse.urlparse(url)
-    if parsed.scheme != "https":
-        raise ValueError(f"Invalid protocol '{parsed.scheme}'. Only HTTPS is allowed.")
+    if "huggingface" not in url:
+        raise ValueError("Only downloads from Hugging Face are allowed")
 
-    hostname = (parsed.hostname or "").lower()
-    if not any(hostname == d or hostname.endswith("." + d) for d in ALLOWED_DOMAINS):
-        raise ValueError("Only downloads from Hugging Face are allowed.")
-
-    return url
-
-
-def get_supported_audio_video_extensions():
     try:
-        subtitle_codecs = set()
-        decoders_res = subprocess.run(
-            ["ffmpeg", "-decoders"], capture_output=True, text=True, check=True
-        )
-        for line in decoders_res.stdout.splitlines():
-            line_str = line.strip()
-            if len(line_str) > 7 and line_str[0] in ("V", "A", "S"):
-                parts = line_str.split()
-                if len(parts) >= 2:
-                    media_type = line_str[0]
-                    codec_name = parts[1].lower()
-                    if media_type == "S":
-                        subtitle_codecs.add(codec_name)
-
-        demuxers_res = subprocess.run(
-            ["ffmpeg", "-demuxers"], capture_output=True, text=True, check=True
-        )
-        extensions = set()
-        for line in demuxers_res.stdout.splitlines():
-            line_str = line.strip()
-            if line_str.startswith("D"):
-                parts = line_str.split(maxsplit=2)
-                if len(parts) >= 2:
-                    demux_names = parts[1].split(",")
-                    description = parts[2].lower() if len(parts) >= 3 else ""
-
-                    if any(
-                        sub_kw in description
-                        for sub_kw in ["subtitle", "caption", "teletext", "lyrics"]
-                    ):
-                        continue
-
-                    for ext in demux_names:
-                        clean_ext = ext.strip().lower()
-                        if clean_ext in subtitle_codecs:
-                            continue
-                        if clean_ext and clean_ext.isalnum():
-                            extensions.add(f".{clean_ext}")
-
-        return sorted(list(extensions))
-    except Exception as e:
-        print(f"Error querying ffmpeg: {e}")
-        return [".mp3", ".wav", ".flac", ".m4a", ".aac", ".ogg", ".opus", ".wma", ".aiff", ".aif", ".alac", ".caf", ".amr"]
-
-
-def ensure_valid_file(url):
-    url = validate_url(url)
-
-    try:
-        request = urllib.request.Request(url, method="HEAD", headers={"User-Agent": "Mozilla/5.0"})
-        with urllib.request.urlopen(request, timeout=15) as response:
-            content_length = response.headers.get("Content-Length")
-
-        if content_length is None:
-            raise ValueError("Unable to read file info from url. The link might be invalid or unreachable.")
+        with urllib.request.urlopen(url) as response:
+            info = response.info()
+            content_length = info.get("Content-Length")
 
         file_size = int(content_length)
-        # print("debug", url, file_size)
-        if file_size > 900000000 and IS_ZERO_GPU:
-            raise ValueError("The file is too large. Max allowed is 900 MB.")
-
-        return file_size
-
-    except urllib.error.HTTPError as err:
-        raise ValueError(
-            f"HTTP Error {err.code} ({err.reason}): The file at the provided URL does not exist, is private, or has been deleted from Hugging Face. Please verify the link."
-        )
+        if file_size > 500000000:
+            raise ValueError("The file is too large. You can only download files up to 500 MB in size.")
 
     except Exception as e:
         raise e
 
+
 def clear_files(directory):
     time.sleep(15)
     print(f"Clearing files: {directory}.")
     shutil.rmtree(directory)
 
 
-def get_my_model(url_data, progress=gr.Progress(track_tqdm=True)):
+def get_my_model(url_data):
 
     if not url_data:
         return None, None
 
     if "," in url_data:
-        a_, b_ = url_data.split(",")
+        a_, b_ = url_data.split()
         a_, b_ = a_.strip().replace("/blob/", "/resolve/"), b_.strip().replace("/blob/", "/resolve/")
     else:
         a_, b_ = url_data.strip().replace("/blob/", "/resolve/"), None
@@ -349,9 +147,12 @@ def get_my_model(url_data, progress=gr.Progress(track_tqdm=True)):
     os.makedirs(directory, exist_ok=True)
 
     try:
+        get_file_size(a_)
+        if b_:
+            get_file_size(b_)
+
         valid_url = [a_] if not b_ else [a_, b_]
         for link in valid_url:
-            ensure_valid_file(link)
             download_manager(
                 url=link,
                 path=directory,
@@ -360,7 +161,6 @@ def get_my_model(url_data, progress=gr.Progress(track_tqdm=True)):
 
         for f in find_files(directory):
             if f.endswith(".zip"):
-                check_model_safety(f)
                 unzip_in_folder(f, directory)
 
         model = None
@@ -369,7 +169,6 @@ def get_my_model(url_data, progress=gr.Progress(track_tqdm=True)):
 
         for ff in end_files:
             if ff.endswith(".pth"):
-                check_model_safety(ff)
                 model = ff
                 gr.Info(f"Model found: {ff}")
             if ff.endswith(".index"):
@@ -395,13 +194,13 @@ def get_my_model(url_data, progress=gr.Progress(track_tqdm=True)):
         t.start()
 
 
-def add_audio_effects(audio_list, type_output):
+def add_audio_effects(audio_list):
     print("Audio effects")
 
     result = []
     for audio_path in audio_list:
         try:
-            output_path = f'{os.path.splitext(audio_path)[0]}_effects.{type_output}'
+            output_path = f'{os.path.splitext(audio_path)[0]}_effects.wav'
 
             # Initialize audio effects plugins
             board = Pedalboard(
@@ -412,23 +211,13 @@ def add_audio_effects(audio_list, type_output):
                  ]
             )
 
-            # Temporary WAV to hold processed data before exporting
-            temp_wav = f'{os.path.splitext(audio_path)[0]}_temp.wav'
-
             with AudioFile(audio_path) as f:
-                with AudioFile(temp_wav, 'w', f.samplerate, f.num_channels) as o:
+                with AudioFile(output_path, 'w', f.samplerate, f.num_channels) as o:
+                    # Read one second of audio at a time, until the file is empty:
                     while f.tell() < f.frames:
                         chunk = f.read(int(f.samplerate))
                         effected = board(chunk, f.samplerate, reset=False)
                         o.write(effected)
-
-            # Convert with pydub to desired output type
-            audio_seg = AudioSegment.from_file(temp_wav, format=type_output)
-            audio_seg.export(output_path, format=type_output, bitrate=("320k" if type_output == "mp3" else None))
-
-            # Clean up temp file
-            os.remove(temp_wav)
-
             result.append(output_path)
         except Exception as e:
             traceback.print_exc()
@@ -438,13 +227,13 @@ def add_audio_effects(audio_list, type_output):
     return result
 
 
-def apply_noisereduce(audio_list, type_output):
+def apply_noisereduce(audio_list):
     # https://github.com/sa-if/Audio-Denoiser
     print("Noice reduce")
 
     result = []
     for audio_path in audio_list:
-        out_path = f"{os.path.splitext(audio_path)[0]}_noisereduce.{type_output}"
+        out_path = f'{os.path.splitext(audio_path)[0]}_noisereduce.wav'
 
         try:
             # Load audio file
@@ -465,7 +254,7 @@ def apply_noisereduce(audio_list, type_output):
             )
 
             # Save reduced audio to file
-            reduced_audio.export(out_path, format=type_output, bitrate=("320k" if type_output == "mp3" else None))
+            reduced_audio.export(out_path, format="wav")
             result.append(out_path)
 
         except Exception as e:
@@ -477,17 +266,13 @@ def apply_noisereduce(audio_list, type_output):
 
 
 @spaces.GPU()
-def convert_now(audio_files, random_tag, converter, type_output, steps):
-    for step in range(steps):
-        audio_files = converter(
-            audio_files,
-            random_tag,
-            overwrite=False,
-            parallel_workers=(2 if IS_COLAB else 8),
-            type_output=type_output,
-        )
-
-    return audio_files
+def convert_now(audio_files, random_tag, converter):
+    return converter(
+        audio_files,
+        random_tag,
+        overwrite=False,
+        parallel_workers=8
+    )
 
 
 def run(
@@ -502,8 +287,6 @@ def run(
     c_b_p,
     active_noise_reduce,
     audio_effects,
-    type_output,
-    steps,
 ):
     if not audio_files:
         raise ValueError("The audio pls")
@@ -517,9 +300,9 @@ def run(
     except Exception as e:
         print(e)
 
-    if file_m is not None and (file_m.endswith(".txt") or file_m.endswith(".zip")):
+    if file_m is not None and file_m.endswith(".txt"):
         file_m, file_index = find_my_model(file_m, file_index)
-        # print(file_m, file_index)
+        print(file_m, file_index)
 
     random_tag = "USER_"+str(random.randint(10000000, 99999999))
 
@@ -533,58 +316,27 @@ def run(
         respiration_median_filtering=r_m_f,
         envelope_ratio=e_r,
         consonant_breath_protection=c_b_p,
-        resample_sr=0,
+        resample_sr=44100 if audio_files[0].endswith('.mp3') else 0, 
     )
     time.sleep(0.1)
 
-    result = convert_now(audio_files, random_tag, converter, type_output, steps)
+    result = convert_now(audio_files, random_tag, converter)
 
     if active_noise_reduce:
-        result = apply_noisereduce(result, type_output)
+        result = apply_noisereduce(result)
 
     if audio_effects:
-        result = add_audio_effects(result, type_output)
+        result = add_audio_effects(result)
 
     return result
 
 
-def clear_player():
-    return None
-
-
-def load_first_audio(output_files, play_audio):
-    if not play_audio or not output_files:
-        return None
-    first_file = output_files[0]
-    if isinstance(first_file, dict):
-        first_file = first_file.get("name")
-    return first_file
-
-
-def audio_source_conf():
-    return gr.Radio(
-        choices=["📁 Upload", "🗣️ TTS", "🎙️ Record"],
-        value="📁 Upload",
-        label="Audio Source",
-    )
-
-
-def mic_conf():
-    return gr.Audio(
-        sources=["microphone"],
-        type="filepath",
-        label="Record Audio",
-        visible=False,
-    )
-
-
 def audio_conf():
     return gr.File(
         label="Audio files",
         file_count="multiple",
         type="filepath",
         container=True,
-        file_types=supported_extensions,
     )
 
 
@@ -593,7 +345,6 @@ def model_conf():
         label="Model file",
         type="filepath",
         height=130,
-        file_types=[".txt", ".pth", ".zip"],
     )
 
 
@@ -624,7 +375,6 @@ def index_conf():
         label="Index file",
         type="filepath",
         height=130,
-        file_types=[".index", ".txt"],
     )
 
 
@@ -683,6 +433,15 @@ def output_conf():
     )
 
 
+def active_tts_conf():
+    return gr.Checkbox(
+        False,
+        label="TTS",
+        # info="",
+        container=False,
+    )
+
+
 def tts_voice_conf():
     return gr.Dropdown(
         label="tts voice",
@@ -710,12 +469,13 @@ def tts_button_conf():
     )
 
 
-def player_conf():
+def tts_play_conf():
     return gr.Checkbox(
         False,
-        label="Audio Player",
+        label="Play",
+        # info="",
         container=False,
-        visible=True,
+        visible=False,
     )
 
 
@@ -723,31 +483,12 @@ def sound_gui():
     return gr.Audio(
         value=None,
         type="filepath",
+        # format="mp3",
         autoplay=True,
         visible=False,
-        interactive=False,
-        label="Audio Player",
     )
 
 
-def steps_conf():
-    return gr.Slider(
-        minimum=1,
-        maximum=3,
-        label="Steps",
-        value=1,
-        step=1,
-        interactive=True,
-    )
-
-
-def format_output_gui():
-    return gr.Dropdown(
-        label="Format output:",
-        choices=["wav", "mp3", "flac"],
-        value="wav",
-    )
-
 def denoise_conf():
     return gr.Checkbox(
         False,
@@ -768,7 +509,7 @@ def effects_conf():
     )
 
 
-def infer_tts_audio(tts_voice, tts_text, play_audio):
+def infer_tts_audio(tts_voice, tts_text, play_tts):
     out_dir = "output"
     folder_tts = "USER_"+str(random.randint(10000, 99999))
 
@@ -777,25 +518,23 @@ def infer_tts_audio(tts_voice, tts_text, play_audio):
     out_path = os.path.join(out_dir, folder_tts, "tts.mp3")
 
     asyncio.run(edge_tts.Communicate(tts_text, "-".join(tts_voice.split('-')[:-1])).save(out_path))
-    player_audio = out_path if play_audio else None
-    return [out_path], player_audio
+    if play_tts:
+        return [out_path], out_path
+    return [out_path], None
 
 
-def show_components_source(mode):
-    return (
-        gr.update(visible=(mode == "🗣️ TTS")),
-        gr.update(visible=(mode == "🗣️ TTS")),
-        gr.update(visible=(mode == "🗣️ TTS")),
-        gr.update(visible=(mode == "🎙️ Record")),
+def show_components_tts(value_active):
+    return gr.update(
+        visible=value_active
+    ), gr.update(
+        visible=value_active
+    ), gr.update(
+        visible=value_active
+    ), gr.update(
+        visible=value_active
     )
 
 
-def sync_mic_audio(mic_path):
-    if not mic_path:
-        return None
-    return [mic_path]
-
-
 def down_active_conf():
     return gr.Checkbox(
         False,
@@ -833,17 +572,12 @@ def show_components_down(value_active):
     )
 
 
-CSS = ""
-supported_extensions = get_supported_audio_video_extensions()
-print("Supported extensions found:", supported_extensions)
-
-
 def get_gui(theme):
-    with gr.Blocks(theme=theme, css=CSS, fill_width=True, fill_height=False, delete_cache=delete_cache_time) as app:
+    with gr.Blocks(theme=theme, delete_cache=(3200, 3200)) as app:
         gr.Markdown(title)
         gr.Markdown(description)
 
-        audio_source = audio_source_conf()
+        active_tts = active_tts_conf()
         with gr.Row():
             with gr.Column(scale=1):
                 tts_text = tts_text_conf()
@@ -852,28 +586,29 @@ def get_gui(theme):
                     with gr.Column():
                         with gr.Row():
                             tts_voice = tts_voice_conf()
+                            tts_active_play = tts_play_conf()
 
                 tts_button = tts_button_conf()
+                tts_play = sound_gui()
+
+        active_tts.change(
+            fn=show_components_tts,
+            inputs=[active_tts],
+            outputs=[tts_voice, tts_text, tts_button, tts_active_play],
+        )
 
-        mic_aud = mic_conf()
         aud = audio_conf()
         # gr.HTML("<hr>")
 
-        audio_source.change(
-            fn=show_components_source,
-            inputs=[audio_source],
-            outputs=[tts_voice, tts_text, tts_button, mic_aud],
-        )
-
-        mic_aud.change(
-            fn=sync_mic_audio,
-            inputs=[mic_aud],
-            outputs=[aud],
+        tts_button.click(
+            fn=infer_tts_audio,
+            inputs=[tts_voice, tts_text, tts_active_play],
+            outputs=[aud, tts_play],
         )
 
         down_active_gui = down_active_conf()
         down_info = gr.Markdown(
-            f"Provide a link to a zip file, like this one: `https://huggingface.co/MrDawg/ToothBrushing/resolve/main/ToothBrushing.zip?download=true`, or separate links with a comma for the .pth and .index files, like this: `{test_model}`",
+            "Provide a link to a zip file, like this one: `https://huggingface.co/mrmocciai/Models/resolve/main/Genshin%20Impact/ayaka-v2.zip?download=true`, or separate links with a comma for the .pth and .index files, like this: `https://huggingface.co/sail-rvc/ayaka-jp/resolve/main/model.pth?download=true, https://huggingface.co/sail-rvc/ayaka-jp/resolve/main/model.index?download=true`",
             visible=False
         )
         with gr.Row():
@@ -899,36 +634,17 @@ def get_gui(theme):
             [model, indx]
         )
 
-        with gr.Accordion(label="Advanced settings", open=False):
-            algo = pitch_algo_conf()
-            algo_lvl = pitch_lvl_conf()
-            indx_inf = index_inf_conf()
-            res_fc = respiration_filter_conf()
-            envel_r = envelope_ratio_conf()
-            const = consonant_protec_conf()
-            steps_gui = steps_conf()
-            format_out = format_output_gui()
-            with gr.Row():
-                with gr.Column():
-                    with gr.Row():
-                        denoise_gui = denoise_conf()
-                        effects_gui = effects_conf()
-                        player_gui = player_conf()
-
-        player_audio = sound_gui()
-
-        player_gui.change(
-            fn=lambda val: gr.update(visible=val),
-            inputs=[player_gui],
-            outputs=[player_audio],
-        )
-
-        tts_button.click(
-            fn=infer_tts_audio,
-            inputs=[tts_voice, tts_text, player_gui],
-            outputs=[aud, player_audio],
-        )
-
+        algo = pitch_algo_conf()
+        algo_lvl = pitch_lvl_conf()
+        indx_inf = index_inf_conf()
+        res_fc = respiration_filter_conf()
+        envel_r = envelope_ratio_conf()
+        const = consonant_protec_conf()
+        with gr.Row():
+            with gr.Column():
+                with gr.Row():
+                    denoise_gui = denoise_conf()
+                    effects_gui = effects_conf()
         button_base = button_conf()
         output_base = output_conf()
 
@@ -946,19 +662,8 @@ def get_gui(theme):
                 const,
                 denoise_gui,
                 effects_gui,
-                format_out,
-                steps_gui,
             ],
             outputs=[output_base],
-        ).success(
-            clear_player,
-            inputs=None,
-            outputs=[player_audio],
-            queue=False,
-        ).success(
-            load_first_audio,
-            inputs=[output_base, player_gui],
-            outputs=[player_audio],
         )
 
         gr.Examples(
@@ -1013,31 +718,24 @@ def get_gui(theme):
             outputs=[output_base],
             cache_examples=False,
         )
-        gr.Markdown(RESOURCES)
 
     return app
 
 
 if __name__ == "__main__":
-    try:
-        tts_voice_list = asyncio.new_event_loop().run_until_complete(get_voices_list(proxy=None))
-        voices = sorted([
-            (" - ".join(reversed(v["FriendlyName"].split("-"))).replace("Microsoft ", "").replace("Online (Natural)", f"({v['Gender']})").strip(), f"{v['ShortName']}-{v['Gender']}")
-            for v in tts_voice_list
-        ])
-    except Exception as e:
-        print(f"Warning: Could not retrieve online voices ({e}). Using default preset.")
-        voices = [("English (United States) - EmmaMultilingual (Female)", "en-US-EmmaMultilingualNeural-Female")]
-    
+
+    tts_voice_list = asyncio.new_event_loop().run_until_complete(edge_tts.list_voices())
+    voices = sorted([f"{v['ShortName']}-{v['Gender']}" for v in tts_voice_list])
+
     app = get_gui(theme)
 
     app.queue(default_concurrency_limit=40)
 
     app.launch(
         max_threads=40,
-        share=IS_COLAB,
+        share=False,
         show_error=True,
         quiet=False,
-        debug=IS_COLAB,
-        ssr_mode=False,
-    )
+        debug=False,
+        allowed_paths=["./downloads/"],
+    )

pre-requirements.txt

@@ -1,2 +0,0 @@
-pip==23.0.1
-Setuptools<=80.6.0

requirements.txt

@@ -1,14 +1,6 @@
-torch==2.9.1
-torchvision==0.24.1
-torchaudio==2.9.1
-infer-rvc-python
-edge_tts==7.2.7
+torch==2.2.0
+infer-rvc-python==1.1.0
+edge-tts
 pedalboard
 noisereduce
-numpy==1.23.5
-transformers<=4.48.3
-# pydantic==2.10.6
-gradio==5.43.1
-spaces
-matplotlib-inline
-picklescan
+numpy==1.23.5