extra-whisper 0.1.0__tar.gz

extra_whisper-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_whisper-0.1.0/PKG-INFO

@@ -0,0 +1,55 @@
+Metadata-Version: 2.1
+Name: extra-whisper
+Version: 0.1.0
+Summary: Extended faster-whisper with yt-dlp remote media transcription
+Author: mohammadmansour200
+Author-email: mohammadamansour03@gmail.com
+Requires-Python: >=3.10
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Requires-Dist: faster-whisper (>=1.2.1)
+Requires-Dist: validators (>=0.35.0,<0.36.0)
+Requires-Dist: yt-dlp (>=2025.10.22)
+Description-Content-Type: text/markdown
+
+<div align="center">
+  <a href="https://pypi.org/project/extra-whisper" target="_blank"><img src="https://img.shields.io/pypi/v/extra-whisper?label=PyPI%20Version&color=limegreen" /></a>
+  <a href="https://pypi.org/project/extra-whisper" target="_blank"><img src="https://img.shields.io/pypi/pyversions/extra-whisper?color=limegreen" /></a>
+  <a href="https://github.com/mohammadmansour200/extra-whisper/blob/master/LICENSE" target="_blank"><img src="https://img.shields.io/pypi/l/extra-whisper?color=limegreen" /></a>
+  <a href="https://pepy.tech/project/extra-whisper" target="_blank"><img src="https://static.pepy.tech/badge/extra-whisper" /></a>
+  <a href="https://baseet.netlify.app/ai" target="_blank"><img src="https://colab.research.google.com/assets/colab-badge.svg" /></a>
+</div>
+
+`extra_whisper`: Extended [faster-whisper](https://github.com/SYSTRAN/faster-whisper) with remote media transcription
+
+## Features
+
+- 🎧 **Transcription** using faster-whisper
+- 📥 **Media download** from URLs (e.g., YouTube) using `yt-dlp`
+- ✅ Local + remote (URL) input support
+
+## Get started
+
+Download package:
+> Requires Python 3.9+
+
+```bash
+pip install extra-whisper
+```
+
+## Usage
+
+```bash
+from extra_whisper.transcribe import extra_transcribe
+
+extra_transcribe(
+    files=[
+        "https://www.youtube.com/watch?v=123",
+        "local_audio.mp3"
+    ],
+    output_dir="outputs"
+)
+
+```

extra_whisper-0.1.0/README.md

@@ -0,0 +1,39 @@
+<div align="center">
+  <a href="https://pypi.org/project/extra-whisper" target="_blank"><img src="https://img.shields.io/pypi/v/extra-whisper?label=PyPI%20Version&color=limegreen" /></a>
+  <a href="https://pypi.org/project/extra-whisper" target="_blank"><img src="https://img.shields.io/pypi/pyversions/extra-whisper?color=limegreen" /></a>
+  <a href="https://github.com/mohammadmansour200/extra-whisper/blob/master/LICENSE" target="_blank"><img src="https://img.shields.io/pypi/l/extra-whisper?color=limegreen" /></a>
+  <a href="https://pepy.tech/project/extra-whisper" target="_blank"><img src="https://static.pepy.tech/badge/extra-whisper" /></a>
+  <a href="https://baseet.netlify.app/ai" target="_blank"><img src="https://colab.research.google.com/assets/colab-badge.svg" /></a>
+</div>
+
+`extra_whisper`: Extended [faster-whisper](https://github.com/SYSTRAN/faster-whisper) with remote media transcription
+
+## Features
+
+- 🎧 **Transcription** using faster-whisper
+- 📥 **Media download** from URLs (e.g., YouTube) using `yt-dlp`
+- ✅ Local + remote (URL) input support
+
+## Get started
+
+Download package:
+> Requires Python 3.9+
+
+```bash
+pip install extra-whisper
+```
+
+## Usage
+
+```bash
+from extra_whisper.transcribe import extra_transcribe
+
+extra_transcribe(
+    files=[
+        "https://www.youtube.com/watch?v=123",
+        "local_audio.mp3"
+    ],
+    output_dir="outputs"
+)
+
+```

extra_whisper-0.1.0/extra_whisper/downloader.py

@@ -0,0 +1,33 @@
+import os
+from typing import Any
+
+import yt_dlp
+
+
+class Downloader:
+    def __init__(self, output_dir: str):
+        self.output_dir = output_dir
+
+    def download(self, url: str) -> str:
+        self._initialize_youtube_dl()
+
+        self.youtube_dl.download(url)
+        url_data = self.youtube_dl.extract_info(url, download=False)
+
+        filename = f"{url_data['id']}.{url_data['ext']}"
+        return filename
+
+    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,
+            'format': "bestaudio"
+        }
+
+        return config

extra_whisper-0.1.0/extra_whisper/transcribe.py

@@ -0,0 +1,96 @@
+import os
+import shutil
+from typing import Optional
+
+import validators
+from faster_whisper import WhisperModel, BatchedInferencePipeline, format_timestamp
+
+from extra_whisper.downloader import Downloader
+
+def extra_transcribe(
+        files: list[str],
+        output_dir: str,
+        model: str = "large-v3",
+        language: Optional[str] = None
+):
+    """
+    Transcripts the list of media files (audio/video), using faster-whisper.
+
+    This function supports both local files and remote URLs (e.g., YouTube links). It handles:
+    - Downloading remote media using yt-dlp
+    - Performing transcription
+    - Cleaning up intermediate files and keeping only the final output
+
+    Parameters:
+        files (list[str]): List of file paths or URLs pointing to audio/video files.
+        output_dir (str): Path to directory where final results will be saved.
+        model (str): Model to be used
+        language (str): Speech language
+
+    Example:
+        extra_transcribe(
+            files=["https://www.youtube.com/watch?v=123", "local_song.mp3"],
+            media_type="audio",
+            quality="medium",
+            output_dir="output"
+        )
+    """
+    abs_output_dir = os.path.abspath(output_dir)
+    temp_output_dir = os.path.join(abs_output_dir, 'tmp')
+
+    processing_files_path = []
+
+    # --- Preparing files for processing ---
+    print("Preparing files...")
+
+    if not files:
+        raise Exception("Please provide files for processing")
+
+    downloader = Downloader(
+        output_dir=temp_output_dir,
+    )
+    for index, url in enumerate(files):
+        is_url = validators.url(url)
+        if is_url:
+            downloaded_file_name = downloader.download(url=url)
+            downloaded_file_path = os.path.join(temp_output_dir, downloaded_file_name)
+
+            processing_files_path.append(downloaded_file_path)
+        else:
+            processing_files_path.append(os.path.abspath(url))
+
+    # --- model inference ---
+    whisper_model = WhisperModel(model, device="cuda", compute_type="float16")
+    batched_model = BatchedInferencePipeline(model=whisper_model)
+
+    # Process each file with batching
+    for file_path in processing_files_path:
+        print(f"Transcribing {file_path}...")
+
+        segments, info = batched_model.transcribe(
+            file_path,
+            batch_size=8,
+            beam_size=5,
+            language=language
+        )
+
+        base_name = os.path.splitext(os.path.basename(file_path))[0]
+        output_file = os.path.join(abs_output_dir, f"{base_name}.srt")
+
+        with open(output_file, 'w', encoding='utf-8') as f:
+            for idx, segment in enumerate(segments, start=1):
+                f.write(f"{idx}\n")
+
+                start_time = format_timestamp(segment.start, always_include_hours=True, decimal_marker=",")
+                end_time = format_timestamp(segment.end, always_include_hours=True, decimal_marker=",")
+                f.write(f"{start_time} --> {end_time}\n")
+
+                f.write(f"{segment.text.strip()}\n")
+                f.write("\n")
+
+        print(f"Saved transcription to {output_file}")
+
+    # --- Cleanup ---
+    if os.path.exists(temp_output_dir):
+        shutil.rmtree(temp_output_dir)
+

extra_whisper-0.1.0/pyproject.toml

@@ -0,0 +1,17 @@
+[tool.poetry]
+name = "extra-whisper"
+version = "0.1.0"
+description = "Extended faster-whisper with yt-dlp remote media transcription"
+authors = ["mohammadmansour200 <mohammadamansour03@gmail.com>"]
+readme = "README.md"
+packages = [{include = "extra_whisper"}]
+
+[tool.poetry.dependencies]
+python = ">=3.10"
+yt-dlp = ">=2025.10.22"
+faster-whisper = ">=1.2.1"
+validators = ">=0.35.0,<0.36.0"
+
+[build-system]
+requires = ["poetry-core"]
+build-backend = "poetry.core.masonry.api"