{ "cells": [ { "cell_type": "markdown", "metadata": { "id": "0pKllbPyK_BC" }, "source": [ "## **Applio NoUI**\n", "A simple, high-quality voice conversion tool focused on ease of use and performance. \n", "\n", "[Support](https://discord.gg/urxFjYmYYh) — [Discord Bot](https://discord.com/oauth2/authorize?client_id=1144714449563955302&permissions=1376674695271&scope=bot%20applications.commands) — [Find Voices](https://applio.org/models) — [GitHub](https://github.com/IAHispano/Applio)\n", "\n", "
\n", "\n", "### **Credits**\n", "- Encryption method: [Hina](https://github.com/hinabl)\n", "- Extra section: [Poopmaster](https://github.com/poiqazwsx)\n", "- Main development: [Applio Team](https://github.com/IAHispano)\n", "- Colab inspired on [RVC v2 Disconnected](https://colab.research.google.com/drive/1XIPCP9ken63S7M6b5ui1b36Cs17sP-NS)." ] }, { "cell_type": "markdown", "metadata": { "id": "Y-iR3WeLMlac" }, "source": [ "### If you restart the runtime, run it again." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "xwZkZGd-H0zT" }, "outputs": [], "source": [ "%cd /content/Applio" ] }, { "cell_type": "markdown", "metadata": { "id": "ymMCTSD6m8qV" }, "source": [ "# Installation\n", "## If the runtime restarts, run the cell above and re-run the installation steps." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "cellView": "form", "id": "yFhAeKGOp9aa" }, "outputs": [], "source": [ "# @title Mount Google Drive\n", "from google.colab import drive\n", "\n", "drive.mount(\"/content/drive\")" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "cellView": "form", "id": "7GysECSxBya4" }, "outputs": [], "source": [ "# @title Clone\n", "!git clone https://github.com/IAHispano/Applio --branch 3.2.7 --single-branch\n", "%cd /content/Applio" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "cellView": "form", "id": "CAXW55BQm0PP" }, "outputs": [], "source": [ "# @title Install\n", "rot_47 = lambda encoded_text: \"\".join(\n", " [\n", " (\n", " chr(\n", " (ord(c) - (ord(\"a\") if c.islower() else ord(\"A\")) - 47) % 26\n", " + (ord(\"a\") if c.islower() else ord(\"A\"))\n", " )\n", " if c.isalpha()\n", " else c\n", " )\n", " for c in encoded_text\n", " ]\n", ")\n", "import codecs\n", "import os\n", "import tarfile\n", "import subprocess\n", "from pathlib import Path\n", "from IPython.display import clear_output\n", "\n", "def vidal_setup(C):\n", " def F():\n", " print(\"Installing pip packages...\")\n", " subprocess.check_call([\"pip\", \"install\", \"-r\", \"requirements.txt\", \"--quiet\"])\n", "\n", " A = \"/content/\" + rot_47(\"Kikpm.ovm.bu\")\n", " D = \"/\"\n", " if not os.path.exists(A):\n", " M = os.path.dirname(A)\n", " os.makedirs(M, exist_ok=True)\n", " print(\"No cached install found..\")\n", " try:\n", " N = codecs.decode(\n", " \"uggcf://uhttvatsnpr.pb/VNUvfcnab/Nccyvb/erfbyir/znva/Raivebzrag/Pbyno/Cache.gne.tm\",\n", " \"rot_13\",\n", " )\n", " subprocess.run([\"wget\", \"-O\", A, N])\n", " print(\"Download completed successfully!\")\n", " except Exception as H:\n", " print(str(H))\n", " if os.path.exists(A):\n", " os.remove(A)\n", " if Path(A).exists():\n", " with tarfile.open(A, \"r:gz\") as I:\n", " I.extractall(D)\n", " print(f\"Extraction of {A} to {D} completed.\")\n", " if os.path.exists(A):\n", " os.remove(A)\n", " if C:\n", " F()\n", " C = False\n", " else:\n", " F()\n", "\n", "\n", "vidal_setup(False)\n", "!pip uninstall torch torchvision torchaudio -y\n", "!pip install torch==2.3.1 torchvision==0.18.1 torchaudio==2.3.1 --upgrade --index-url https://download.pytorch.org/whl/cu121\n", "clear_output()\n", "print(\"Finished installing requirements!\")" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "cellView": "form", "id": "QlTibPnjmj6-" }, "outputs": [], "source": [ "# @title Download models\n", "!python core.py \"prerequisites\" --models \"True\" --exe \"True\" --pretraineds_v1_f0 \"False\" --pretraineds_v2_f0 \"True\" --pretraineds_v1_nof0 \"False\" --pretraineds_v2_nof0 \"False\" " ] }, { "cell_type": "markdown", "metadata": { "id": "YzaeMYsUE97Y" }, "source": [ "# Infer\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "cellView": "form", "id": "v0EgikgjFCjE" }, "outputs": [], "source": [ "# @title Download model\n", "# @markdown Hugging Face or Google Drive\n", "model_link = \"https://huggingface.co/Darwin/Darwin/resolve/main/Darwin.zip\" # @param {type:\"string\"}\n", "\n", "!python core.py download --model_link \"{model_link}\"" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "cellView": "form", "id": "lrCKEOzvDPRu" }, "outputs": [], "source": [ "# @title Run Inference\n", "# @markdown Please upload the audio file to your Google Drive path `/content/drive/MyDrive` and specify its name here. For the model name, use the zip file name without the extension. Alternatively, you can check the path `/content/Applio/logs` for the model name (name of the folder).\n", "\n", "import os\n", "\n", "current_dir = os.getcwd()\n", "\n", "model_name = \"Darwin\" # @param {type:\"string\"}\n", "model_folder = os.path.join(current_dir, f\"logs/{model_name}\")\n", "\n", "if not os.path.exists(model_folder):\n", " raise FileNotFoundError(f\"Model directory not found: {model_folder}\")\n", "\n", "files_in_folder = os.listdir(model_folder)\n", "pth_path = next((f for f in files_in_folder if f.endswith(\".pth\")), None)\n", "index_file = next((f for f in files_in_folder if f.endswith(\".index\")), None)\n", "\n", "if pth_path is None or index_file is None:\n", " raise FileNotFoundError(\"No model found.\")\n", "\n", "pth_file = os.path.join(model_folder, pth_path)\n", "index_file = os.path.join(model_folder, index_file)\n", "\n", "input_path = \"/content/example.wav\" # @param {type:\"string\"}\n", "output_path = \"/content/output.wav\"\n", "export_format = \"WAV\" # @param ['WAV', 'MP3', 'FLAC', 'OGG', 'M4A'] {allow-input: false}\n", "f0_method = \"rmvpe\" # @param [\"crepe\", \"crepe-tiny\", \"rmvpe\", \"fcpe\", \"hybrid[rmvpe+fcpe]\"] {allow-input: false}\n", "f0_up_key = 0 # @param {type:\"slider\", min:-24, max:24, step:0}\n", "filter_radius = 3 # @param {type:\"slider\", min:0, max:10, step:0}\n", "rms_mix_rate = 0.8 # @param {type:\"slider\", min:0.0, max:1.0, step:0.1}\n", "protect = 0.5 # @param {type:\"slider\", min:0.0, max:0.5, step:0.1}\n", "index_rate = 0.7 # @param {type:\"slider\", min:0.0, max:1.0, step:0.1}\n", "hop_length = 128 # @param {type:\"slider\", min:1, max:512, step:0}\n", "clean_strength = 0.7 # @param {type:\"slider\", min:0.0, max:1.0, step:0.1}\n", "split_audio = False # @param{type:\"boolean\"}\n", "clean_audio = False # @param{type:\"boolean\"}\n", "f0_autotune = False # @param{type:\"boolean\"}\n", "formant_shift = False # @param{type:\"boolean\"}\n", "formant_qfrency = 1.0 # @param {type:\"slider\", min:1.0, max:16.0, step:0.1}\n", "formant_timbre = 1.0 # @param {type:\"slider\", min:1.0, max:16.0, step:0.1}\n", "embedder_model = \"contentvec\" # @param [\"contentvec\", \"chinese-hubert-base\", \"japanese-hubert-base\", \"korean-hubert-base\", \"custom\"] {allow-input: false}\n", "embedder_model_custom = \"\" # @param {type:\"string\"}\n", "upscale_audio = False # @param{type:\"boolean\"}\n", "\n", "\n", "# Post-processing effects\n", "if \"post_process\" not in globals():\n", " post_process = False \n", "if \"reverb\" not in globals():\n", " reverb = False \n", "if \"pitch_shift\" not in globals():\n", " pitch_shift = False \n", "if \"limiter\" not in globals():\n", " limiter = False \n", "if \"gain\" not in globals():\n", " gain = False \n", "if \"distortion\" not in globals():\n", " distortion = False \n", "if \"chorus\" not in globals():\n", " chorus = False \n", "if \"bitcrush\" not in globals():\n", " bitcrush = False\n", "if \"clipping\" not in globals():\n", " clipping = False \n", "if \"compressor\" not in globals():\n", " compressor = False \n", "if \"delay\" not in globals():\n", " delay = False\n", "\n", "if \"reverb_room_size\" not in globals():\n", " reverb_room_size = 0.5 \n", "if \"reverb_damping\" not in globals():\n", " reverb_damping = 0.5 \n", "if \"reverb_wet_gain\" not in globals():\n", " reverb_wet_gain = 0.0 \n", "if \"reverb_dry_gain\" not in globals():\n", " reverb_dry_gain = 0.0 \n", "if \"reverb_width\" not in globals():\n", " reverb_width = 1.0 \n", "if \"reverb_freeze_mode\" not in globals():\n", " reverb_freeze_mode = 0.0 \n", "\n", "if \"pitch_shift_semitones\" not in globals():\n", " pitch_shift_semitones = 0.0 \n", "\n", "if \"limiter_threshold\" not in globals():\n", " limiter_threshold = -1.0 \n", "if \"limiter_release_time\" not in globals():\n", " limiter_release_time = 0.05 \n", "\n", "if \"gain_db\" not in globals():\n", " gain_db = 0.0 \n", "\n", "if \"distortion_gain\" not in globals():\n", " distortion_gain = 0.0 \n", "\n", "if \"chorus_rate\" not in globals():\n", " chorus_rate = 1.5 \n", "if \"chorus_depth\" not in globals():\n", " chorus_depth = 0.1 \n", "if \"chorus_center_delay\" not in globals():\n", " chorus_center_delay = 15.0 \n", "if \"chorus_feedback\" not in globals():\n", " chorus_feedback = 0.25 \n", "if \"chorus_mix\" not in globals():\n", " chorus_mix = 0.5 \n", "\n", "if \"bitcrush_bit_depth\" not in globals():\n", " bitcrush_bit_depth = 4 \n", "\n", "if \"clipping_threshold\" not in globals():\n", " clipping_threshold = 0.5 \n", "\n", "if \"compressor_threshold\" not in globals():\n", " compressor_threshold = -20.0\n", "if \"compressor_ratio\" not in globals():\n", " compressor_ratio = 4.0 \n", "if \"compressor_attack\" not in globals():\n", " compressor_attack = 0.001 \n", "if \"compressor_release\" not in globals():\n", " compressor_release = 0.1 \n", "\n", "if \"delay_seconds\" not in globals():\n", " delay_seconds = 0.1\n", "if \"delay_feedback\" not in globals():\n", " delay_feedback = 0.5 \n", "if \"delay_mix\" not in globals():\n", " delay_mix = 0.5 \n", " \n", "!python core.py infer --pitch \"{f0_up_key}\" --filter_radius \"{filter_radius}\" --volume_envelope \"{rms_mix_rate}\" --index_rate \"{index_rate}\" --hop_length \"{hop_length}\" --protect \"{protect}\" --f0_autotune \"{f0_autotune}\" --f0_method \"{f0_method}\" --input_path \"{input_path}\" --output_path \"{output_path}\" --pth_path \"{pth_file}\" --index_path \"{index_file}\" --split_audio \"{split_audio}\" --clean_audio \"{clean_audio}\" --clean_strength \"{clean_strength}\" --export_format \"{export_format}\" --embedder_model \"{embedder_model}\" --embedder_model_custom \"{embedder_model_custom}\" --upscale_audio \"{upscale_audio}\" --formant_shifting \"{formant_shift}\" --formant_qfrency \"{formant_qfrency}\" --formant_timbre \"{formant_timbre}\" --post_process \"{post_process}\" --reverb \"{reverb}\" --pitch_shift \"{pitch_shift}\" --limiter \"{limiter}\" --gain \"{gain}\" --distortion \"{distortion}\" --chorus \"{chorus}\" --bitcrush \"{bitcrush}\" --clipping \"{clipping}\" --compressor \"{compressor}\" --delay \"{delay}\" --reverb_room_size \"{reverb_room_size}\" --reverb_damping \"{reverb_damping}\" --reverb_wet_gain \"{reverb_wet_gain}\" --reverb_dry_gain \"{reverb_dry_gain}\" --reverb_width \"{reverb_width}\" --reverb_freeze_mode \"{reverb_freeze_mode}\" --pitch_shift_semitones \"{pitch_shift_semitones}\" --limiter_threshold \"{limiter_threshold}\" --limiter_release_time \"{limiter_release_time}\" --gain_db \"{gain_db}\" --distortion_gain \"{distortion_gain}\" --chorus_rate \"{chorus_rate}\" --chorus_depth \"{chorus_depth}\" --chorus_center_delay \"{chorus_center_delay}\" --chorus_feedback \"{chorus_feedback}\" --chorus_mix \"{chorus_mix}\" --bitcrush_bit_depth \"{bitcrush_bit_depth}\" --clipping_threshold \"{clipping_threshold}\" --compressor_threshold \"{compressor_threshold}\" --compressor_ratio \"{compressor_ratio}\" --compressor_attack \"{compressor_attack}\" --compressor_release \"{compressor_release}\" --delay_seconds \"{delay_seconds}\" --delay_feedback \"{delay_feedback}\" --delay_mix \"{delay_mix}\"\n", "\n", "from IPython.display import Audio, display, clear_output\n", "\n", "output_path = output_path.replace(\".wav\", f\".{export_format.lower()}\")\n", "# clear_output()\n", "display(Audio(output_path, autoplay=True))" ] }, { "cell_type": "markdown", "metadata": { "id": "yrWw2h9d2TRn" }, "source": [ "## **Advanced Settings**" ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "cellView": "form", "id": "J43qejJ-2Tpp" }, "outputs": [], "source": [ "# @title # Post-processing effects\n", "post_process = False # @param{type:\"boolean\"}\n", "reverb = False # @param{type:\"boolean\"}\n", "pitch_shift = False # @param{type:\"boolean\"}\n", "limiter = False # @param{type:\"boolean\"}\n", "gain = False # @param{type:\"boolean\"}\n", "distortion = False # @param{type:\"boolean\"}\n", "chorus = False # @param{type:\"boolean\"}\n", "bitcrush = False # @param{type:\"boolean\"}\n", "clipping = False # @param{type:\"boolean\"}\n", "compressor = False # @param{type:\"boolean\"}\n", "delay = False # @param{type:\"boolean\"}\n", "\n", "reverb_room_size = 0.5 # @param {type:\"slider\", min:0.0, max:1.0, step:0.1}\n", "reverb_damping = 0.5 # @param {type:\"slider\", min:0.0, max:1.0, step:0.1}\n", "reverb_wet_gain = 0.0 # @param {type:\"slider\", min:-20.0, max:20.0, step:0.1}\n", "reverb_dry_gain = 0.0 # @param {type:\"slider\", min:-20.0, max:20.0, step:0.1}\n", "reverb_width = 1.0 # @param {type:\"slider\", min:0.0, max:1.0, step:0.1}\n", "reverb_freeze_mode = 0.0 # @param {type:\"slider\", min:0.0, max:1.0, step:0.1}\n", "\n", "pitch_shift_semitones = 0.0 # @param {type:\"slider\", min:-12.0, max:12.0, step:0.1}\n", "\n", "limiter_threshold = -1.0 # @param {type:\"slider\", min:-20.0, max:0.0, step:0.1}\n", "limiter_release_time = 0.05 # @param {type:\"slider\", min:0.0, max:1.0, step:0.01}\n", "\n", "gain_db = 0.0 # @param {type:\"slider\", min:-20.0, max:20.0, step:0.1}\n", "\n", "distortion_gain = 0.0 # @param {type:\"slider\", min:0.0, max:1.0, step:0.1}\n", "\n", "chorus_rate = 1.5 # @param {type:\"slider\", min:0.1, max:10.0, step:0.1}\n", "chorus_depth = 0.1 # @param {type:\"slider\", min:0.0, max:1.0, step:0.1}\n", "chorus_center_delay = 15.0 # @param {type:\"slider\", min:0.0, max:50.0, step:0.1}\n", "chorus_feedback = 0.25 # @param {type:\"slider\", min:0.0, max:1.0, step:0.1}\n", "chorus_mix = 0.5 # @param {type:\"slider\", min:0.0, max:1.0, step:0.1}\n", "\n", "bitcrush_bit_depth = 4 # @param {type:\"slider\", min:1, max:16, step:1}\n", "\n", "clipping_threshold = 0.5 # @param {type:\"slider\", min:0.0, max:1.0, step:0.1}\n", "\n", "compressor_threshold = -20.0 # @param {type:\"slider\", min:-60.0, max:0.0, step:0.1}\n", "compressor_ratio = 4.0 # @param {type:\"slider\", min:1.0, max:20.0, step:0.1}\n", "compressor_attack = 0.001 # @param {type:\"slider\", min:0.0, max:0.1, step:0.001}\n", "compressor_release = 0.1 # @param {type:\"slider\", min:0.0, max:1.0, step:0.01}\n", "\n", "delay_seconds = 0.1 # @param {type:\"slider\", min:0.0, max:1.0, step:0.01}\n", "delay_feedback = 0.5 # @param {type:\"slider\", min:0.0, max:1.0, step:0.1}\n", "delay_mix = 0.5 # @param {type:\"slider\", min:0.0, max:1.0, step:0.1}\n" ] }, { "cell_type": "markdown", "metadata": { "id": "1QkabnLlF2KB" }, "source": [ "# Train" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "cellView": "form", "id": "oBzqm4JkGGa0" }, "outputs": [], "source": [ "# @title Preprocess Dataset\n", "import os\n", "os.environ['TF_CPP_MIN_LOG_LEVEL'] = '3'\n", "model_name = \"Darwin\" # @param {type:\"string\"}\n", "dataset_path = \"/content/drive/MyDrive/Darwin_Dataset\" # @param {type:\"string\"}\n", "\n", "sample_rate = \"40k\" # @param [\"32k\", \"40k\", \"48k\"] {allow-input: false}\n", "sr = int(sample_rate.rstrip(\"k\")) * 1000\n", "cpu_cores = 2 # @param {type:\"slider\", min:1, max:2, step:1}\n", "cut_preprocess = True # @param{type:\"boolean\"}\n", "process_effects = False # @param{type:\"boolean\"}\n", "noise_reduction = False # @param{type:\"boolean\"}\n", "noise_reduction_strength = 0.7 # @param {type:\"slider\", min:0.0, max:1.0, step:0.1}\n", "\n", "!python core.py preprocess --model_name \"{model_name}\" --dataset_path \"{dataset_path}\" --sample_rate \"{sr}\" --cpu_cores \"{cpu_cores}\" --cut_preprocess \"{cut_preprocess}\" --process_effects \"{process_effects}\" --noise_reduction \"{noise_reduction}\" --noise_reduction_strength \"{noise_reduction_strength}\"" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "cellView": "form", "id": "zWMiMYfRJTJv" }, "outputs": [], "source": [ "# @title Extract Features\n", "rvc_version = \"v2\" # @param [\"v2\", \"v1\"] {allow-input: false}\n", "f0_method = \"rmvpe\" # @param [\"crepe\", \"crepe-tiny\", \"rmvpe\"] {allow-input: false}\n", "hop_length = 128 # @param {type:\"slider\", min:1, max:512, step:0}\n", "\n", "sr = int(sample_rate.rstrip(\"k\")) * 1000\n", "cpu_cores = 2 # @param {type:\"slider\", min:1, max:2, step:1}\n", "embedder_model = \"contentvec\" # @param [\"contentvec\", \"chinese-hubert-base\", \"japanese-hubert-base\", \"korean-hubert-base\", \"custom\"] {allow-input: false}\n", "embedder_model_custom = \"\" # @param {type:\"string\"}\n", "\n", "!python core.py extract --model_name \"{model_name}\" --rvc_version \"{rvc_version}\" --f0_method \"{f0_method}\" --hop_length \"{hop_length}\" --sample_rate \"{sr}\" --cpu_cores \"{cpu_cores}\" --gpu \"0\" --embedder_model \"{embedder_model}\" --embedder_model_custom \"{embedder_model_custom}\"" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "cellView": "form", "id": "TI6LLdIzKAIa" }, "outputs": [], "source": [ "# @title Train\n", "import threading\n", "import time\n", "import os\n", "import shutil\n", "import hashlib\n", "import time\n", "\n", "LOGS_FOLDER = \"/content/Applio/logs/\"\n", "GOOGLE_DRIVE_PATH = \"/content/drive/MyDrive/RVC_Backup\"\n", "\n", "\n", "def import_google_drive_backup():\n", " print(\"Importing Google Drive backup...\")\n", " for root, dirs, files in os.walk(GOOGLE_DRIVE_PATH):\n", " for filename in files:\n", " filepath = os.path.join(root, filename)\n", " if os.path.isfile(filepath):\n", " backup_filepath = os.path.join(\n", " LOGS_FOLDER, os.path.relpath(filepath, GOOGLE_DRIVE_PATH)\n", " )\n", " backup_folderpath = os.path.dirname(backup_filepath)\n", " if not os.path.exists(backup_folderpath):\n", " os.makedirs(backup_folderpath)\n", " print(f\"Created backup folder: {backup_folderpath}\", flush=True)\n", " shutil.copy2(filepath, backup_filepath)\n", " print(f\"Imported file from Google Drive backup: {filename}\")\n", " print(\"Google Drive backup import completed.\")\n", "\n", "\n", "def get_md5_hash(file_path):\n", " hash_md5 = hashlib.md5()\n", " with open(file_path, \"rb\") as f:\n", " for chunk in iter(lambda: f.read(4096), b\"\"):\n", " hash_md5.update(chunk)\n", " return hash_md5.hexdigest()\n", "\n", "\n", "if \"autobackups\" not in globals():\n", " autobackups = False\n", "# @markdown ### 💾 AutoBackup\n", "cooldown = 15 # @param {type:\"slider\", min:0, max:100, step:0}\n", "auto_backups = True # @param{type:\"boolean\"}\n", "def backup_files():\n", " print(\"\\nStarting backup loop...\")\n", " last_backup_timestamps_path = os.path.join(\n", " LOGS_FOLDER, \"last_backup_timestamps.txt\"\n", " )\n", " fully_updated = False\n", "\n", " while True:\n", " try:\n", " updated_files = 0\n", " deleted_files = 0\n", " new_files = 0\n", " last_backup_timestamps = {}\n", "\n", " try:\n", " with open(last_backup_timestamps_path, \"r\") as f:\n", " last_backup_timestamps = dict(line.strip().split(\":\") for line in f)\n", " except FileNotFoundError:\n", " pass\n", "\n", " for root, dirs, files in os.walk(LOGS_FOLDER):\n", " # Excluding \"zips\" and \"mute\" directories\n", " if \"zips\" in dirs:\n", " dirs.remove(\"zips\")\n", " if \"mute\" in dirs:\n", " dirs.remove(\"mute\")\n", "\n", " for filename in files:\n", " if filename != \"last_backup_timestamps.txt\":\n", " filepath = os.path.join(root, filename)\n", " if os.path.isfile(filepath):\n", " backup_filepath = os.path.join(\n", " GOOGLE_DRIVE_PATH,\n", " os.path.relpath(filepath, LOGS_FOLDER),\n", " )\n", " backup_folderpath = os.path.dirname(backup_filepath)\n", " if not os.path.exists(backup_folderpath):\n", " os.makedirs(backup_folderpath)\n", " last_backup_timestamp = last_backup_timestamps.get(filepath)\n", " current_timestamp = os.path.getmtime(filepath)\n", " if (\n", " last_backup_timestamp is None\n", " or float(last_backup_timestamp) < current_timestamp\n", " ):\n", " shutil.copy2(filepath, backup_filepath)\n", " last_backup_timestamps[filepath] = str(current_timestamp)\n", " if last_backup_timestamp is None:\n", " new_files += 1\n", " else:\n", " updated_files += 1\n", "\n", "\n", " for filepath in list(last_backup_timestamps.keys()):\n", " if not os.path.exists(filepath):\n", " backup_filepath = os.path.join(\n", " GOOGLE_DRIVE_PATH, os.path.relpath(filepath, LOGS_FOLDER)\n", " )\n", " if os.path.exists(backup_filepath):\n", " os.remove(backup_filepath)\n", " deleted_files += 1\n", " del last_backup_timestamps[filepath]\n", "\n", "\n", " if updated_files > 0 or deleted_files > 0 or new_files > 0:\n", " print(f\"Backup Complete: {new_files} new, {updated_files} updated, {deleted_files} deleted.\")\n", " fully_updated = False\n", " elif not fully_updated:\n", " print(\"Files are up to date.\")\n", " fully_updated = True\n", "\n", " with open(last_backup_timestamps_path, \"w\") as f:\n", " for filepath, timestamp in last_backup_timestamps.items():\n", " f.write(f\"{filepath}:{timestamp}\\n\")\n", "\n", " time.sleep(cooldown if fully_updated else 0.1)\n", "\n", "\n", " except Exception as error:\n", " print(f\"An error occurred during backup: {error}\")\n", "\n", "\n", "if autobackups:\n", " autobackups = False\n", " print(\"Autobackup Disabled\")\n", "else:\n", " autobackups = True\n", " print(\"Autobackup Enabled\") \n", "# @markdown ### ⚙️ Train Settings\n", "total_epoch = 800 # @param {type:\"integer\"}\n", "batch_size = 15 # @param {type:\"slider\", min:1, max:25, step:0}\n", "gpu = 0\n", "sr = int(sample_rate.rstrip(\"k\")) * 1000\n", "pretrained = True # @param{type:\"boolean\"}\n", "cleanup = False # @param{type:\"boolean\"}\n", "cache_data_in_gpu = False # @param{type:\"boolean\"}\n", "tensorboard = True # @param{type:\"boolean\"}\n", "# @markdown ### ➡️ Choose how many epochs your model will be stored\n", "save_every_epoch = 10 # @param {type:\"slider\", min:1, max:100, step:0}\n", "save_only_latest = False # @param{type:\"boolean\"}\n", "save_every_weights = False # @param{type:\"boolean\"}\n", "overtraining_detector = False # @param{type:\"boolean\"}\n", "overtraining_threshold = 50 # @param {type:\"slider\", min:1, max:100, step:0}\n", "# @markdown ### ❓ Optional\n", "# @markdown In case you select custom pretrained, you will have to download the pretraineds and enter the path of the pretraineds.\n", "custom_pretrained = False # @param{type:\"boolean\"}\n", "g_pretrained_path = \"/content/Applio/rvc/models/pretraineds/pretraineds_custom/G48k.pth\" # @param {type:\"string\"}\n", "d_pretrained_path = \"/content/Applio/rvc/models/pretraineds/pretraineds_custom/D48k.pth\" # @param {type:\"string\"}\n", "\n", "if \"pretrained\" not in globals():\n", " pretrained = True\n", "\n", "if \"custom_pretrained\" not in globals():\n", " custom_pretrained = False\n", "\n", "if \"g_pretrained_path\" not in globals():\n", " g_pretrained_path = \"Custom Path\"\n", "\n", "if \"d_pretrained_path\" not in globals():\n", " d_pretrained_path = \"Custom Path\"\n", "\n", "\n", "def start_train():\n", " if tensorboard == True:\n", " %load_ext tensorboard\n", " %tensorboard --logdir /content/Applio/logs/\n", " !python core.py train --model_name \"{model_name}\" --rvc_version \"{rvc_version}\" --save_every_epoch \"{save_every_epoch}\" --save_only_latest \"{save_only_latest}\" --save_every_weights \"{save_every_weights}\" --total_epoch \"{total_epoch}\" --sample_rate \"{sr}\" --batch_size \"{batch_size}\" --gpu \"{gpu}\" --pretrained \"{pretrained}\" --custom_pretrained \"{custom_pretrained}\" --g_pretrained_path \"{g_pretrained_path}\" --d_pretrained_path \"{d_pretrained_path}\" --overtraining_detector \"{overtraining_detector}\" --overtraining_threshold \"{overtraining_threshold}\" --cleanup \"{cleanup}\" --cache_data_in_gpu \"{cache_data_in_gpu}\"\n", "\n", "\n", "server_thread = threading.Thread(target=start_train)\n", "server_thread.start()\n", "\n", "if auto_backups:\n", " backup_files()\n", "else:\n", " while True:\n", " time.sleep(10)" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "cellView": "form", "id": "bHLs5AT4Q1ck" }, "outputs": [], "source": [ "# @title Generate index file\n", "index_algorithm = \"Auto\" # @param [\"Auto\", \"Faiss\", \"KMeans\"] {allow-input: false}\n", "!python core.py index --model_name \"{model_name}\" --rvc_version \"{rvc_version}\" --index_algorithm \"{index_algorithm}\"" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "cellView": "form", "id": "X_eU_SoiHIQg" }, "outputs": [], "source": [ "# @title Save model\n", "# @markdown Enter the name of the model and the steps. You can find it in your `/content/Applio/logs` folder.\n", "%cd /content\n", "import os, shutil, sys\n", "\n", "model_name = \"Darwin\" # @param {type:\"string\"}\n", "model_epoch = 800 # @param {type:\"integer\"}\n", "save_big_file = False # @param {type:\"boolean\"}\n", "\n", "if os.path.exists(\"/content/zips\"):\n", " shutil.rmtree(\"/content/zips\")\n", "print(\"Removed zips.\")\n", "\n", "os.makedirs(f\"/content/zips/{model_name}/\", exist_ok=True)\n", "print(\"Created zips.\")\n", "\n", "logs_folder = f\"/content/Applio/logs/{model_name}/\"\n", "weight_file = None\n", "if not os.path.exists(logs_folder):\n", " print(f\"Model folder not found.\")\n", " sys.exit(\"\")\n", "\n", "for filename in os.listdir(logs_folder):\n", " if filename.startswith(f\"{model_name}_{model_epoch}e\") and filename.endswith(\".pth\"):\n", " weight_file = filename\n", " break\n", "if weight_file is None:\n", " print(\"There is no weight file with that name\")\n", " sys.exit(\"\")\n", "if not save_big_file:\n", " !cp {logs_folder}added_*.index /content/zips/{model_name}/\n", " !cp {logs_folder}total_*.npy /content/zips/{model_name}/\n", " !cp {logs_folder}{weight_file} /content/zips/{model_name}/\n", " %cd /content/zips\n", " !zip -r {model_name}.zip {model_name}\n", "if save_big_file:\n", " %cd /content/Applio\n", " latest_steps = -1\n", " logs_folder = \"./logs/\" + model_name\n", " for filename in os.listdir(logs_folder):\n", " if filename.startswith(\"G_\") and filename.endswith(\".pth\"):\n", " steps = int(filename.split(\"_\")[1].split(\".\")[0])\n", " if steps > latest_steps:\n", " latest_steps = steps\n", " MODELZIP = model_name + \".zip\"\n", " !mkdir -p /content/zips\n", " ZIPFILEPATH = os.path.join(\"/content/zips\", MODELZIP)\n", " for filename in os.listdir(logs_folder):\n", " if \"G_\" in filename or \"D_\" in filename:\n", " if str(latest_steps) in filename:\n", " !zip -r {ZIPFILEPATH} {os.path.join(logs_folder, filename)}\n", " else:\n", " !zip -r {ZIPFILEPATH} {os.path.join(logs_folder, filename)}\n", "\n", "!mkdir -p /content/drive/MyDrive/RVC_Backup/\n", "shutil.move(\n", " f\"/content/zips/{model_name}.zip\",\n", " f\"/content/drive/MyDrive/RVC_Backup/{model_name}.zip\",\n", ")\n", "%cd /content/Applio\n", "shutil.rmtree(\"/content/zips\")" ] }, { "cell_type": "markdown", "metadata": { "id": "OaKoymXsyEYN" }, "source": [ "# Resume-training" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "cellView": "form", "id": "d3KgLAYnyHkP" }, "outputs": [], "source": [ "# @title Load a Backup\n", "from google.colab import drive\n", "import os\n", "import shutil\n", "\n", "# @markdown Put the exact name you put as your Model Name in Applio.\n", "modelname = \"My-Project\" # @param {type:\"string\"}\n", "source_path = \"/content/drive/MyDrive/RVC_Backup/\" + modelname\n", "destination_path = \"/content/Applio/logs/\" + modelname\n", "backup_timestamps_file = \"last_backup_timestamps.txt\"\n", "if not os.path.exists(source_path):\n", " print(\n", " \"The model folder does not exist. Please verify the name is correct or check your Google Drive.\"\n", " )\n", "else:\n", " time_ = os.path.join(\"/content/drive/MyDrive/RVC_Backup/\", backup_timestamps_file)\n", " time__ = os.path.join(\"/content/Applio/logs/\", backup_timestamps_file)\n", " if os.path.exists(time_):\n", " shutil.copy(time_, time__)\n", " shutil.copytree(source_path, destination_path)\n", " print(\"Model backup loaded successfully.\")" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "cellView": "form", "id": "sc9DzvRCyJ2d" }, "outputs": [], "source": [ "# @title Set training variables\n", "# @markdown ### ➡️ Use the same as you did previously\n", "model_name = \"Darwin\" # @param {type:\"string\"}\n", "sample_rate = \"40k\" # @param [\"32k\", \"40k\", \"48k\"] {allow-input: false}\n", "rvc_version = \"v2\" # @param [\"v2\", \"v1\"] {allow-input: false}\n", "f0_method = \"rmvpe\" # @param [\"crepe\", \"crepe-tiny\", \"rmvpe\"] {allow-input: false}\n", "hop_length = 128 # @param {type:\"slider\", min:1, max:512, step:0}\n", "sr = int(sample_rate.rstrip(\"k\")) * 1000" ] } ], "metadata": { "accelerator": "GPU", "colab": { "collapsed_sections": [ "ymMCTSD6m8qV" ], "provenance": [], "toc_visible": true }, "kernelspec": { "display_name": "Python 3", "name": "python3" }, "language_info": { "name": "python" } }, "nbformat": 4, "nbformat_minor": 0 }