extra-roformers 0.1.0__tar.gz

extra_roformers-0.1.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2025-present, Mohammad Abdul-Tawab.
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

extra_roformers-0.1.0/PKG-INFO

@@ -0,0 +1,72 @@
+Metadata-Version: 2.4
+Name: extra-roformers
+Version: 0.1.0
+Summary: Extended audio-seprator with yt-dlp and Video support.
+License: MIT
+License-File: LICENSE
+Keywords: audio-separation,vocal-extract,music-source-separation,yt-dlp,ffmpeg,deep-learning,music-remove
+Author: mohammadmansour200
+Author-email: mohammadamansour03@gmail.com
+Requires-Python: >=3.10
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
+Requires-Dist: audio-separator[gpu] (>=0.30.0)
+Requires-Dist: validators (>=0.35.0)
+Requires-Dist: yt-dlp (>=2026.2.21)
+Description-Content-Type: text/markdown
+
+<div align="center">
+  <a href="https://pypi.org/project/extra-roformers" target="_blank"><img src="https://img.shields.io/pypi/v/extra-roformers?label=PyPI%20Version&color=limegreen" /></a>
+  <a href="https://pypi.org/project/extra-roformers" target="_blank"><img src="https://img.shields.io/pypi/pyversions/extra-roformers?color=limegreen" /></a>
+  <a href="https://github.com/mohammadmansour200/extra-roformers/blob/main/LICENSE" target="_blank"><img src="https://img.shields.io/pypi/l/extra-roformers?color=limegreen" /></a>
+  <a href="https://pepy.tech/project/extra-roformers" target="_blank"><img src="https://static.pepy.tech/badge/extra-roformers" /></a>
+  <a href="https://baseet.netlify.app/ai" target="_blank"><img src="https://colab.research.google.com/assets/colab-badge.svg" /></a>
+</div>
+
+`extra-roformers`: Extended [audio-separator](https://github.com/nomadkaraoke/python-audio-separator/) with yt-dlp media downloading and Video
+Music removal
+
+## Features
+
+- 🎧 **Vocal isolation** using Mel-Band RoFormer model
+- 📥 **Media download** from URLs (e.g., YouTube) using `yt-dlp`
+- 📁 Works with both **audio** and **video** files
+- ✅ Local + remote (URL) input support
+
+## Get started
+
+*Make sure you have [ffmpeg](https://www.ffmpeg.org/download.html) installed.*
+
+```bash
+sudo apt install ffmpeg
+```
+
+Download package:
+> Requires Python 3.10+
+
+```bash
+pip install extra-roformers
+```
+
+## Usage
+
+```bash
+from extra_roformers.separate import extra_separator
+
+extra_separator(
+    files=[
+        "https://www.youtube.com/watch?v=123",
+        "local_audio.mp3"
+    ],
+    download_format="audio",   # or "video"
+    quality="medium",     # "low", "medium", "high"
+    output_dir="outputs"
+)
+
+```
+

extra_roformers-0.1.0/README.md

@@ -0,0 +1,49 @@
+<div align="center">
+  <a href="https://pypi.org/project/extra-roformers" target="_blank"><img src="https://img.shields.io/pypi/v/extra-roformers?label=PyPI%20Version&color=limegreen" /></a>
+  <a href="https://pypi.org/project/extra-roformers" target="_blank"><img src="https://img.shields.io/pypi/pyversions/extra-roformers?color=limegreen" /></a>
+  <a href="https://github.com/mohammadmansour200/extra-roformers/blob/main/LICENSE" target="_blank"><img src="https://img.shields.io/pypi/l/extra-roformers?color=limegreen" /></a>
+  <a href="https://pepy.tech/project/extra-roformers" target="_blank"><img src="https://static.pepy.tech/badge/extra-roformers" /></a>
+  <a href="https://baseet.netlify.app/ai" target="_blank"><img src="https://colab.research.google.com/assets/colab-badge.svg" /></a>
+</div>
+
+`extra-roformers`: Extended [audio-separator](https://github.com/nomadkaraoke/python-audio-separator/) with yt-dlp media downloading and Video
+Music removal
+
+## Features
+
+- 🎧 **Vocal isolation** using Mel-Band RoFormer model
+- 📥 **Media download** from URLs (e.g., YouTube) using `yt-dlp`
+- 📁 Works with both **audio** and **video** files
+- ✅ Local + remote (URL) input support
+
+## Get started
+
+*Make sure you have [ffmpeg](https://www.ffmpeg.org/download.html) installed.*
+
+```bash
+sudo apt install ffmpeg
+```
+
+Download package:
+> Requires Python 3.10+
+
+```bash
+pip install extra-roformers
+```
+
+## Usage
+
+```bash
+from extra_roformers.separate import extra_separator
+
+extra_separator(
+    files=[
+        "https://www.youtube.com/watch?v=123",
+        "local_audio.mp3"
+    ],
+    download_format="audio",   # or "video"
+    quality="medium",     # "low", "medium", "high"
+    output_dir="outputs"
+)
+
+```

extra_roformers-0.1.0/pyproject.toml

@@ -0,0 +1,32 @@
+[project]
+name = "extra-roformers"
+version = "0.1.0"
+description = "Extended audio-seprator with yt-dlp and Video support."
+license = "MIT License"
+packages = [{ include = "extra_roformers", from = "src" }]
+include = ["LICENSE"]
+readme = "README.md"
+repository = "https://github.com/mohammadmansour200/extra_roformers"
+homepage = "https://baseet.netlify.app/ai"
+keywords = [
+    "audio-separation",
+    "vocal-extract",
+    "music-source-separation",
+    "yt-dlp",
+    "ffmpeg",
+    "deep-learning",
+    "music-remove"
+]
+authors = [
+    { name = "mohammadmansour200", email = "mohammadamansour03@gmail.com" }
+]
+requires-python = ">=3.10"
+dependencies = [
+    "yt-dlp>=2026.2.21",
+    "validators>=0.35.0",
+    "audio-separator[gpu]>=0.30.0",
+]
+
+[build-system]
+requires = ["poetry-core>=2.0.0,<3.0.0"]
+build-backend = "poetry.core.masonry.api"

extra_roformers-0.1.0/src/extra_roformers/downloader.py

@@ -0,0 +1,50 @@
+import os
+from typing import Any
+
+import yt_dlp
+
+int_quality = {
+    "high": "192",
+    "medium": "128",
+    "low": "96"
+}
+
+
+class Downloader:
+    def __init__(self, output_dir: str, media_type: str, quality: str):
+        self.output_dir = output_dir
+        self.media_type = media_type
+        self.quality = quality
+        self.is_output_video = self.media_type == "video"
+
+    def download(self, url: str) -> tuple[str, Any]:
+        self._initialize_youtube_dl()
+
+        url_data = self.youtube_dl.extract_info(url, download=True)
+
+        full_path = self.youtube_dl.prepare_filename(url_data)
+
+        return full_path
+
+    def _initialize_youtube_dl(self) -> None:
+        self.youtube_dl = yt_dlp.YoutubeDL(self._config())
+
+    def _config(self) -> dict[str, Any]:
+        config = {
+            'ignoreerrors': True,
+            'noplaylist': True,
+            'outtmpl': os.path.join(self.output_dir, '%(id)s.%(ext)s'),
+            'quiet': True,
+            'verbose': False,
+        }
+
+        is_high_quality = self.quality == "high"
+
+        if not self.is_output_video:
+            config['format'] = f'bestaudio[abr>={int_quality[self.quality]}]/bestaudio'
+
+        if self.is_output_video and not is_high_quality:
+            config[
+                'format'] = 'bestvideo[height<=720]+bestaudio' if self.quality == "medium" else 'bestvideo[height<=360]+bestaudio'
+
+        return config

extra_roformers-0.1.0/src/extra_roformers/ffmpeg_utils.py

@@ -0,0 +1,79 @@
+import os
+import shutil
+import subprocess
+
+
+class FFMPEGUtils:
+    """
+    FFmpeg utils used in separation.
+
+    When created, FFMPEG binary path will be checked,
+    raising exception if not found. Such path could be inferred using
+    `FFMPEG_PATH` environment variable.
+    """
+
+    def __init__(self) -> None:
+        """
+        Default constructor, ensure FFMPEG binaries are available.
+
+        Raises:
+            ValueError:
+                If ffmpeg or ffprobe is not found.
+        """
+        for binary in ("ffmpeg", "ffprobe"):
+            if shutil.which(binary) is None:
+                raise Exception("ffmpeg_utils:{} binary not found".format(binary))
+
+    def replace_video_audio(self, input_video_path: str, input_audio_path: str, final_output_path: str):
+        try:
+            subprocess.run([
+                "ffmpeg",
+                "-y",
+                "-loglevel", "quiet",
+                "-an",
+                "-i", input_video_path,
+                "-i", input_audio_path,
+                "-map", "0:v",
+                "-map", "1:a",
+                "-c:v", "copy",
+                "-c:a", "copy",
+                final_output_path, ], check=True)
+        except Exception as e:
+            raise Exception(
+                "ffmpeg_utils:An error occurred with ffmpeg (see ffmpeg output below)\n\n{}".format(
+                    e
+                )
+            )
+
+    def is_video(self, path: str) -> bool:
+        try:
+            result = subprocess.run(
+                ['ffprobe', '-v', 'error', '-select_streams', 'v:0',
+                 '-show_entries', 'stream=codec_type', '-of', 'csv=p=0', path],
+                stdout=subprocess.PIPE,
+                stderr=subprocess.PIPE,
+                text=True, check=True
+            )
+            return 'video' in result.stdout
+        except Exception as e:
+            return False
+
+    def convert_to_audio_format(self, output_dir: str, output_name: str, audio_format: str):
+        processed_demucs_file_path = os.path.join(output_dir, f"{output_name}.wav")
+        final_output_file_path = os.path.join(output_dir,
+                                              f"{output_name}.{'mp3' if audio_format is None else audio_format}")
+        try:
+            subprocess.run([
+                "ffmpeg",
+                "-y",
+                "-loglevel", "quiet",
+                "-i", processed_demucs_file_path,
+                final_output_file_path
+            ]
+                , check=True)
+        except Exception as e:
+            raise Exception(
+                "ffmpeg_utils:An error occurred with ffmpeg (see ffmpeg output below)\n\n{}".format(
+                    e
+                )
+            )

extra_roformers-0.1.0/src/extra_roformers/separate.py

@@ -0,0 +1,167 @@
+import os
+import shutil
+
+import validators
+from audio_separator.separator import Separator
+
+from extra_roformers.downloader import Downloader
+from extra_roformers.ffmpeg_utils import FFMPEGUtils
+
+video_audio_track_ext_map = {
+    ".mp4": "aac",
+    ".m4v": "aac",
+    ".webm": "opus",
+    ".flv": "aac",
+    ".mkv": "opus",
+    ".mov": "aac",
+    ".avi": "mp3",
+    ".ts": "aac",
+    ".ogg": "vorbis"
+}
+
+i18n = {
+    "en": {
+        "preparing": "Preparing files...",
+        "error_no_files": "Please provide files for processing.",
+        "saving_video": "Saving video in {path}",
+        "saving_audio": "Saving audio in {path}"
+    },
+    "ar": {
+        "preparing": "جاري تحضير الملفات من أجل معالجتها...",
+        "error_no_files": "لا يوجد ملفات للمعالجة.",
+        "saving_video": "جاري حفظ الفيديو في {path}",
+        "saving_audio": "جاري حفظ الصوت في {path}"
+    }
+}
+
+def get_output_format(input_filename):
+    ext = Path(input_filename).suffix.lower()
+
+    if ext in video_audio_track_ext_map:
+        target_format = video_audio_track_ext_map[ext]
+    elif ext in ['.wav', '.flac']:
+        target_format = ext.replace('.', '')
+    else:
+        target_format = "mp3"
+
+    return target_format
+
+def extra_separator(
+        files: list[str],
+        download_format: str,
+        quality: str,
+        output_dir: str,
+        model: str = 'vocals_mel_band_roformer.ckpt',
+        lang: str = "ar"
+):
+    """
+    Separates vocals from a list of media files (audio/video), using audio-separator, and replaces
+    the audio track in video files with the extracted vocals if file was a video.
+
+    This function supports both local files and remote URLs (e.g., YouTube links). It handles:
+    - Downloading remote media using yt-dlp
+    - Performing source separation using Demucs (vocal isolation)
+    - For video: replacing the original audio with vocals using ffmpeg
+    - Cleaning up intermediate files and keeping only the final output
+    - Segmented processing in case of low-end devices
+
+    Parameters:
+        files (list[str]): List of file paths or URLs pointing to audio/video files.
+        download_format (str): Either "audio" or "video". Determines post-processing behavior.
+        quality (str): Quality level for yt-dlp downloading ("high", "low", "medium").
+        output_dir (str): Path to directory where final results will be saved.
+        model (str): model to be used
+        lang (str): Language for logs
+
+    Notes:
+        - Requires `ffmpeg`, and internet access for remote URLs.
+        - For audio files, only the isolated vocal track is kept in mp3 format.
+        - For video files, a new `.mp4` is generated with vocals replacing original audio.
+        - Temporary files are stored in a `tmp/` subfolder inside the output directory and deleted after completion.
+
+    Example:
+        extra_separator(
+            files=["https://www.youtube.com/watch?v=123", "local_song.mp3"],
+            media_type="audio",
+            quality="medium",
+            output_dir="output"
+        )
+    """
+    t = i18n[lang]
+
+    abs_output_dir = os.path.abspath(output_dir)
+    temp_output_dir = os.path.join(abs_output_dir, 'tmp')
+    separator_output_dir = os.path.join(abs_output_dir, model)
+
+    # Create directories if they don't exist
+    os.makedirs(temp_output_dir, exist_ok=True)
+    os.makedirs(separator_output_dir, exist_ok=True)
+
+    ffmpeg_utils = FFMPEGUtils()
+
+    file_to_be_processed = []
+
+    # --- Preparing files for processing ---
+    print(t["preparing"])
+
+    if not files:
+        raise Exception(t["error_no_files"])
+
+    downloader = Downloader(
+        output_dir=temp_output_dir,
+        media_type=download_format,
+        quality=quality
+    )
+    for index, url in enumerate(files):
+        is_url = validators.url(url)
+        if is_url:
+            downloaded_file_path = downloader.download(url=url)
+
+            file_to_be_processed.append(downloaded_file_path)
+        else:
+            file_to_be_processed.append(os.path.abspath(url))
+
+    # --- Demucs model inference ---
+    separator = Separator(
+        output_dir=separator_output_dir,
+        output_single_stem='Vocals',
+    )
+    separator.load_model(model)
+
+    # --- Postprocess ---
+    for original_file_path in files_to_be_processed:
+        file_path_obj = Path(original_file_path)
+        original_file_ext = file_path_obj.suffix.lower()
+        vocal_output_name = file_path_obj.stem
+
+        # --- Separate Audio ---
+        target_format = get_output_format(original_file_path)
+        separated_files = separator.separate(path, output_format=target_format.upper(), output_filename=f"{path_name_only}_vocals")
+
+        vocal_file_name = separated_files[0]
+        vocal_output_path = os.path.join(separator_output_dir, vocal_file_name)
+
+        # --- Post-Processing ---
+        is_video = ffmpeg_utils.is_video(original_file_path)
+
+        if is_video:
+            print(t["saving_video"].format(path=abs_output_dir))
+
+            final_video_output_path = os.path.join(abs_output_dir, f"{vocal_output_name}{original_file_ext}")
+
+            ffmpeg_utils.replace_video_audio(
+                input_video_path=original_file_path,
+                input_audio_path=vocal_output_path,
+                final_output_path=final_video_output_path
+            )
+        else:
+            print(t["saving_video"].format(path=abs_output_dir))
+
+            final_audio_output_path = os.path.join(abs_output_dir, vocal_file_name)
+            shutil.move(vocal_output_path, final_audio_output_path)
+
+    # --- Cleanup ---
+    if os.path.exists(temp_output_dir):
+        shutil.rmtree(temp_output_dir)
+    if os.path.exists(demucs_output_dir):
+        shutil.rmtree(demucs_output_dir)