VidLingua 0.1.0a1__tar.gz

vidlingua-0.1.0a1/PKG-INFO

@@ -0,0 +1,83 @@
+Metadata-Version: 2.4
+Name: VidLingua
+Version: 0.1.0a1
+Summary: A command-line tool that automates the translation and dubbing of YouTube videos.
+Home-page: https://github.com/vprayag2005/VidLingua
+Author: vprayag2005
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Topic :: Multimedia :: Video
+Classifier: Programming Language :: Python :: 3
+Classifier: Operating System :: OS Independent
+Requires-Python: >=3.8
+Description-Content-Type: text/markdown
+Requires-Dist: yt-dlp
+Requires-Dist: faster-whisper
+Requires-Dist: deep-translator
+Requires-Dist: edge-tts
+Dynamic: author
+Dynamic: classifier
+Dynamic: description
+Dynamic: description-content-type
+Dynamic: home-page
+Dynamic: requires-dist
+Dynamic: requires-python
+Dynamic: summary
+
+# VidLingua
+
+VidLingua is a command-line tool that automates the translation and dubbing of YouTube videos. It downloads the original video, transcribes the audio, translates the text into a target language, generates a new voiceover using neural text-to-speech, and merges the new audio back into the video file.
+
+## Dependencies
+
+- **Python 3.8+**
+- **FFmpeg**: Must be installed and accessible in your system's `PATH`.
+
+## Installation
+
+1. Clone or download this repository.
+2. Navigate to the `VidLingua` directory.
+3. Install the package using `pip`:
+
+```bash
+pip install -e .
+```
+
+## Usage
+
+```bash
+Vidlingua run --yt-url <URL> --translate-to <LANG_CODE> [OPTIONS]
+```
+
+### Options
+
+| Option | Description | Default |
+| :--- | :--- | :--- |
+| `--yt-url` | **(Required)** The YouTube video URL. | |
+| `--translate-to` | **(Required)** The target language code (e.g., `en`, `hi`, `ml`). | |
+| `--destination` | Directory to save the final video. | `~/Downloads` |
+
+### Examples
+
+Translate to English and save to the default Downloads folder:
+```bash
+Vidlingua run --yt-url "https://youtube.com/watch?v=..." --translate-to "en"
+```
+
+Translate to Hindi and save to a specific directory:
+```bash
+Vidlingua run --yt-url "https://youtube.com/watch?v=..." --translate-to "hi" --destination "./my_videos"
+```
+
+## Supported Languages
+
+VidLingua maps high-quality Neural TTS voices to the following language codes. Other language codes supported by Google Translate will fall back to a default voice.
+
+| Code | Language |
+| :--- | :--- |
+| `en` | English |
+| `hi` | Hindi |
+| `ml` | Malayalam |
+| `ta` | Tamil |
+| `kn` | Kannada |
+| `te` | Telugu |

vidlingua-0.1.0a1/README.md

@@ -0,0 +1,57 @@
+# VidLingua
+
+VidLingua is a command-line tool that automates the translation and dubbing of YouTube videos. It downloads the original video, transcribes the audio, translates the text into a target language, generates a new voiceover using neural text-to-speech, and merges the new audio back into the video file.
+
+## Dependencies
+
+- **Python 3.8+**
+- **FFmpeg**: Must be installed and accessible in your system's `PATH`.
+
+## Installation
+
+1. Clone or download this repository.
+2. Navigate to the `VidLingua` directory.
+3. Install the package using `pip`:
+
+```bash
+pip install -e .
+```
+
+## Usage
+
+```bash
+Vidlingua run --yt-url <URL> --translate-to <LANG_CODE> [OPTIONS]
+```
+
+### Options
+
+| Option | Description | Default |
+| :--- | :--- | :--- |
+| `--yt-url` | **(Required)** The YouTube video URL. | |
+| `--translate-to` | **(Required)** The target language code (e.g., `en`, `hi`, `ml`). | |
+| `--destination` | Directory to save the final video. | `~/Downloads` |
+
+### Examples
+
+Translate to English and save to the default Downloads folder:
+```bash
+Vidlingua run --yt-url "https://youtube.com/watch?v=..." --translate-to "en"
+```
+
+Translate to Hindi and save to a specific directory:
+```bash
+Vidlingua run --yt-url "https://youtube.com/watch?v=..." --translate-to "hi" --destination "./my_videos"
+```
+
+## Supported Languages
+
+VidLingua maps high-quality Neural TTS voices to the following language codes. Other language codes supported by Google Translate will fall back to a default voice.
+
+| Code | Language |
+| :--- | :--- |
+| `en` | English |
+| `hi` | Hindi |
+| `ml` | Malayalam |
+| `ta` | Tamil |
+| `kn` | Kannada |
+| `te` | Telugu |

vidlingua-0.1.0a1/VidLingua.egg-info/PKG-INFO

@@ -0,0 +1,83 @@
+Metadata-Version: 2.4
+Name: VidLingua
+Version: 0.1.0a1
+Summary: A command-line tool that automates the translation and dubbing of YouTube videos.
+Home-page: https://github.com/vprayag2005/VidLingua
+Author: vprayag2005
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Topic :: Multimedia :: Video
+Classifier: Programming Language :: Python :: 3
+Classifier: Operating System :: OS Independent
+Requires-Python: >=3.8
+Description-Content-Type: text/markdown
+Requires-Dist: yt-dlp
+Requires-Dist: faster-whisper
+Requires-Dist: deep-translator
+Requires-Dist: edge-tts
+Dynamic: author
+Dynamic: classifier
+Dynamic: description
+Dynamic: description-content-type
+Dynamic: home-page
+Dynamic: requires-dist
+Dynamic: requires-python
+Dynamic: summary
+
+# VidLingua
+
+VidLingua is a command-line tool that automates the translation and dubbing of YouTube videos. It downloads the original video, transcribes the audio, translates the text into a target language, generates a new voiceover using neural text-to-speech, and merges the new audio back into the video file.
+
+## Dependencies
+
+- **Python 3.8+**
+- **FFmpeg**: Must be installed and accessible in your system's `PATH`.
+
+## Installation
+
+1. Clone or download this repository.
+2. Navigate to the `VidLingua` directory.
+3. Install the package using `pip`:
+
+```bash
+pip install -e .
+```
+
+## Usage
+
+```bash
+Vidlingua run --yt-url <URL> --translate-to <LANG_CODE> [OPTIONS]
+```
+
+### Options
+
+| Option | Description | Default |
+| :--- | :--- | :--- |
+| `--yt-url` | **(Required)** The YouTube video URL. | |
+| `--translate-to` | **(Required)** The target language code (e.g., `en`, `hi`, `ml`). | |
+| `--destination` | Directory to save the final video. | `~/Downloads` |
+
+### Examples
+
+Translate to English and save to the default Downloads folder:
+```bash
+Vidlingua run --yt-url "https://youtube.com/watch?v=..." --translate-to "en"
+```
+
+Translate to Hindi and save to a specific directory:
+```bash
+Vidlingua run --yt-url "https://youtube.com/watch?v=..." --translate-to "hi" --destination "./my_videos"
+```
+
+## Supported Languages
+
+VidLingua maps high-quality Neural TTS voices to the following language codes. Other language codes supported by Google Translate will fall back to a default voice.
+
+| Code | Language |
+| :--- | :--- |
+| `en` | English |
+| `hi` | Hindi |
+| `ml` | Malayalam |
+| `ta` | Tamil |
+| `kn` | Kannada |
+| `te` | Telugu |

vidlingua-0.1.0a1/VidLingua.egg-info/SOURCES.txt

@@ -0,0 +1,17 @@
+README.md
+setup.py
+VidLingua.egg-info/PKG-INFO
+VidLingua.egg-info/SOURCES.txt
+VidLingua.egg-info/dependency_links.txt
+VidLingua.egg-info/entry_points.txt
+VidLingua.egg-info/requires.txt
+VidLingua.egg-info/top_level.txt
+Vidlingua/__init__.py
+Vidlingua/cli.py
+Vidlingua/core.py
+Vidlingua/exceptions.py
+Vidlingua.egg-info/PKG-INFO
+Vidlingua.egg-info/SOURCES.txt
+Vidlingua.egg-info/dependency_links.txt
+Vidlingua.egg-info/entry_points.txt
+Vidlingua.egg-info/top_level.txt

vidlingua-0.1.0a1/VidLingua.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

vidlingua-0.1.0a1/VidLingua.egg-info/entry_points.txt

@@ -0,0 +1,2 @@
+[console_scripts]
+Vidlingua = Vidlingua.cli:main

vidlingua-0.1.0a1/VidLingua.egg-info/requires.txt

@@ -0,0 +1,4 @@
+yt-dlp
+faster-whisper
+deep-translator
+edge-tts

vidlingua-0.1.0a1/VidLingua.egg-info/top_level.txt

@@ -0,0 +1 @@
+Vidlingua

vidlingua-0.1.0a1/Vidlingua/__init__.py

@@ -0,0 +1 @@
+# Vidlingua Package

vidlingua-0.1.0a1/Vidlingua/cli.py

@@ -0,0 +1,67 @@
+import argparse
+import os
+import sys
+import tempfile
+
+from Vidlingua.core import (
+    download_media,
+    speech_to_text,
+    translate_text,
+    text_to_audio,
+    merge_audio_video,
+)
+from Vidlingua.exceptions import VidLinguaError
+
+
+def main():
+    parser = argparse.ArgumentParser(
+        description="VidLingua: Video Translation and Dubbing"
+    )
+    subparsers = parser.add_subparsers(dest="command", required=True)
+
+    run_parser = subparsers.add_parser("run", help="Run the translation process")
+    run_parser.add_argument("--yt-url", required=True, help="YouTube video URL")
+    run_parser.add_argument(
+        "--translate-to",
+        required=True,
+        help="Language code to translate to (e.g., 'en', 'es', 'hi')",
+    )
+    run_parser.add_argument(
+        "--destination",
+        default=os.path.expanduser("~/Downloads"),
+        help="Directory to save the final dubbed video (defaults to system Downloads folder)",
+    )
+
+    args = parser.parse_args()
+
+    if args.command == "run":
+        video_url = args.yt_url
+        translate_lang = args.translate_to
+
+        try:
+            with tempfile.TemporaryDirectory() as temp_dir:
+                audio_file, video_file = download_media(video_url, temp_dir)
+                transcribed_text = speech_to_text(audio_file)
+
+                if translate_lang.strip():
+                    lang_code = translate_lang.strip().replace("'", "").replace('"', "")
+                    final_translated_text = translate_text(transcribed_text, lang_code)
+
+                    if final_translated_text:
+                        temp_audio_path = os.path.join(temp_dir, "translated_audio.mp3")
+                        generated_audio_path = text_to_audio(
+                            final_translated_text, lang_code, temp_audio_path
+                        )
+                        if generated_audio_path:
+                            os.makedirs(args.destination, exist_ok=True)
+                            final_video_path = os.path.join(
+                                args.destination, "final_dubbed_video.mp4"
+                            )
+                            merge_audio_video(video_file, generated_audio_path, final_video_path)
+        except VidLinguaError as e:
+            print(f"\n[Error] {e}")
+            sys.exit(1)
+
+
+if __name__ == "__main__":
+    main()

vidlingua-0.1.0a1/Vidlingua/core.py

@@ -0,0 +1,130 @@
+import asyncio
+import os
+import subprocess
+
+import edge_tts
+import yt_dlp
+from deep_translator import GoogleTranslator
+from faster_whisper import WhisperModel
+
+from Vidlingua.exceptions import (
+    AudioGenerationError,
+    DownloadError,
+    MergeError,
+    TranslationError,
+)
+
+VOICE_MAPPING = {
+    "en": "en-US-ChristopherNeural",
+    "hi": "hi-IN-MadhurNeural",
+    "ml": "ml-IN-MidhunNeural",
+    "ta": "ta-IN-PallaviNeural",
+    "kn": "kn-IN-GaganNeural",
+    "te": "te-IN-MohanNeural",
+}
+
+
+def get_youtube_options(format_str, outtmpl):
+    """Returns the yt-dlp configuration options."""
+    opts = {
+        "format": format_str,
+        "outtmpl": outtmpl,
+        "extractor_args": {"youtube": {"player_client": ["android", "web"]}},
+    }
+    return opts
+
+
+def download_media(url, temp_dir):
+    """
+    Downloads both the audio (for transcription)
+    and the video without audio (for dubbing).
+    """
+
+    audio_opts = get_youtube_options(
+        "bestaudio/best", os.path.join(temp_dir, "original_audio.%(ext)s")
+    )
+    video_opts = get_youtube_options(
+        "bestvideo/best", os.path.join(temp_dir, "original_video.%(ext)s")
+    )
+
+    try:
+        with yt_dlp.YoutubeDL(audio_opts) as ydl:
+            info_dict = ydl.extract_info(url, download=True)
+            audio_path = ydl.prepare_filename(info_dict)
+
+        with yt_dlp.YoutubeDL(video_opts) as ydl:
+            info_dict = ydl.extract_info(url, download=True)
+            video_path = ydl.prepare_filename(info_dict)
+
+        return audio_path, video_path
+    except yt_dlp.utils.DownloadError as e:
+        raise DownloadError(f"Failed to download media: {e}")
+
+
+def speech_to_text(audio_path):
+    model = WhisperModel("base", compute_type="int8", cpu_threads=8)
+
+    segments, info = model.transcribe(audio_path, beam_size=1, vad_filter=True)
+
+    text = "".join([segment.text for segment in segments])
+
+    return text.strip()
+
+
+def translate_text(text, target_lang):
+    try:
+        translator = GoogleTranslator(source="auto", target=target_lang)
+
+        chunk_size = 4500
+        chunks = [text[i : i + chunk_size] for i in range(0, len(text), chunk_size)]
+
+        translated_text = ""
+        for i, chunk in enumerate(chunks):
+            translated_text += translator.translate(chunk) + " "
+
+        return translated_text.strip()
+    except Exception as e:
+        raise TranslationError(f"Failed to translate text: {e}")
+
+
+def text_to_audio(text, lang_code, output_path):
+    try:
+        voice = VOICE_MAPPING.get(lang_code.lower(), "en-US-ChristopherNeural")
+
+        async def _generate():
+            communicate = edge_tts.Communicate(text, voice)
+            await communicate.save(output_path)
+
+        asyncio.run(_generate())
+
+        return output_path
+    except Exception as e:
+        raise AudioGenerationError(f"Failed to generate audio: {e}")
+
+
+def merge_audio_video(video_path, audio_path, output_path="final_dubbed_video.mp4"):
+    command = [
+        "ffmpeg",
+        "-y",
+        "-i",
+        video_path,
+        "-i",
+        audio_path,
+        "-map",
+        "0:v:0",
+        "-map",
+        "1:a:0",
+        "-c:v",
+        "copy",
+        "-c:a",
+        "aac",
+        "-shortest",
+        output_path,
+    ]
+    try:
+        subprocess.run(
+            command, check=True, stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL
+        )
+        return output_path
+    except subprocess.CalledProcessError as e:
+        raise MergeError("Failed to merge audio and video. Ensure FFmpeg is installed.")

vidlingua-0.1.0a1/Vidlingua/exceptions.py

@@ -0,0 +1,28 @@
+class VidLinguaError(Exception):
+    """Base exception for all VidLingua errors."""
+
+    pass
+
+
+class DownloadError(VidLinguaError):
+    """Raised when media downloading fails."""
+
+    pass
+
+
+class TranslationError(VidLinguaError):
+    """Raised when text translation fails."""
+
+    pass
+
+
+class AudioGenerationError(VidLinguaError):
+    """Raised when generating audio from text fails."""
+
+    pass
+
+
+class MergeError(VidLinguaError):
+    """Raised when merging audio and video fails."""
+
+    pass

vidlingua-0.1.0a1/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

vidlingua-0.1.0a1/setup.py

@@ -0,0 +1,36 @@
+import os
+from setuptools import setup, find_packages
+
+# Read the contents of your README file
+with open("README.md", "r", encoding="utf-8") as fh:
+    long_description = fh.read()
+
+setup(
+    name="VidLingua",
+    version="0.1.0a1",  # 'a1' indicates Alpha 1
+    author="vprayag2005",
+    description="A command-line tool that automates the translation and dubbing of YouTube videos.",
+    long_description=long_description,
+    long_description_content_type="text/markdown",
+    url="https://github.com/vprayag2005/VidLingua",
+    packages=find_packages(),
+    install_requires=[
+        "yt-dlp",
+        "faster-whisper",
+        "deep-translator",
+        "edge-tts",
+    ],
+    entry_points={
+        "console_scripts": [
+            "Vidlingua=Vidlingua.cli:main",
+        ],
+    },
+    classifiers=[
+        "Development Status :: 3 - Alpha",
+        "Intended Audience :: Developers",
+        "Topic :: Multimedia :: Video",
+        "Programming Language :: Python :: 3",
+        "Operating System :: OS Independent",
+    ],
+    python_requires=">=3.8",
+)