Add audio transcription script using Faster Whisper

Transcribing/readme.md

@@ -0,0 +1,74 @@
+# MUST READ BEFORE EXECUTION (ONE TIME ONLY)
+
+This script allows you to **transcribe audio files (MP3, WAV, etc.) to text** using the **Faster Whisper** Python library. Please read these instructions carefully before running the script.  
+
+---
+
+## 1️⃣ Prerequisites
+
+1. **Install Python (3.8 or higher)**  
+   Download and install from [https://www.python.org/downloads/](https://www.python.org/downloads/).  
+   Make sure to **add Python to your system PATH** during installation.
+   OR you can simply run this command in terminal to install python: "winget install Python.Python.3"
+   for MAC: brew install python
+
+2. **Install ffmpeg** (required for audio decoding)  
+   - **Windows:** [https://ffmpeg.org/download.html](https://ffmpeg.org/download.html)  
+   - **Linux:** `sudo apt install ffmpeg`  
+   - **Mac:** `brew install ffmpeg`  
+   Make sure ffmpeg is **added to your system PATH** so the script can find it. (Explanation in End on how to do this)
+    Now you can verify if both are installed by writing "python --version" or "py --version" and "ffmpeg -version"
+---
+
+## 2️⃣ Audio File Requirements
+
+- Place the audio file you want to transcribe **in the same folder** as the script.  
+- Make sure to use the **correct file extension** (e.g., `.mp3`, `.wav`). 
+- Also works for videos (.mp4)
+- The output transcription will also be saved in the same folder with the filename:  
+
+
+---
+
+## 3️⃣ Choosing the Model
+
+The script supports the following **Faster Whisper models**. Please choose the one that fits your needs:
+
+| Model   | Speed & Size                 | Accuracy                     |
+|---------|------------------------------|-------------------------------|
+| tiny    | Fastest, smallest            | Least accurate               |
+| small   | Fast, small                  | Slightly more accurate       |
+| base    | Moderate speed & size        | Good accuracy                |
+| medium  | Slower, larger               | More accurate                |
+| large   | Slowest, largest             | Most accurate                |
+
+**Tip:** Use `base` for a good balance between speed and accuracy. Model names are **case-insensitive**.
+
+---
+
+## 4️⃣ Running the Script
+
+1. Open a terminal or command prompt in the folder containing the script.  
+2. Run the script: type "python transcription.py"
+3. Follow Along the input instructions prompted on screen
+
+## 5 Setting ffmpeg in Your System PATH
+
+To make ffmpeg accessible to the script:
+
+**Windows:**  
+1. Download and unzip ffmpeg.  
+2. Copy the path to the `bin` folder (e.g., `C:\ffmpeg\bin`).  
+3. Open **Environment Variables → Path → Edit → New**, paste the path, and click **OK**.  
+4. Restart the terminal or command prompt.
+
+**Linux / Mac:**  
+write export PATH=$PATH:/path/to/ffmpeg/bin
+
+## Developer
+- GitHub: https://github.com/Syed-56
+- LinkedIn: https://www.linkedin.com/in/syedsultan10
+
+## Dependencies & Licenses
+- Faster Whisper: MIT License (https://github.com/guillaumekln/faster-whisper)
+- ffmpeg: LGPL/GPL License (https://ffmpeg.org)

Transcribing/transcription.py

@@ -0,0 +1,55 @@
+import sys
+import subprocess
+import shutil
+
+#functions to isntall python packages
+def installPackage(pkgName):
+    print(f"[INFO] {pkgName} not found. Installing...")
+    subprocess.check_call([sys.executable, "-m", "pip", "install", pkgName])
+    print(f"[INFO] {pkgName} installed successfully!")
+
+try:
+    from faster_whisper import WhisperModel
+except ImportError:
+    installPackage("faster_whisper")
+    from faster_whisper import WhisperModel
+
+# Check for ffmpeg
+if shutil.which("ffmpeg") is None:
+    print(
+        "[ERROR] ffmpeg not found on your system.\n"
+        "Please install ffmpeg to use this script:\n"
+        "- Windows: https://ffmpeg.org/download.html\n"
+        "- Linux: sudo apt install ffmpeg\n"
+        "- Mac: brew install ffmpeg"
+    )
+    sys.exit(1)
+
+print("[INFO] All dependencies are ready!")
+
+audio_file = input("Enter the path to your audio file (MP3, WAV, MP4 etc.): ").strip()
+valid_models = ["tiny", "small", "base", "medium", "large"]
+print("Models (Sorted from Fastest-Smallest-Least Accurate To Slowest-Largest-Most Accurate):-\n")
+print("Make Sure to write .en after writing model name if you only want english transcription (Less work for model)\n")
+print(valid_models)
+model_size = input("\nEnter the model size: ").strip().lower()
+
+if model_size not in valid_models:
+    print(f"[ERROR] Invalid model size '{model_size}'. Please choose from {valid_models}.")
+    sys.exit(1)
+
+model = WhisperModel(model_size)
+
+import os
+if not os.path.isfile(audio_file):
+    print(f"[ERROR] File '{audio_file}' does not exist. Please check the path and try again.")
+    sys.exit(1)
+
+output_file = audio_file.rsplit(".", 1)[0] + "_transcript.txt"
+segments, info = model.transcribe(audio_file)
+
+with open(output_file, "w", encoding="utf-8") as f:
+    for segment in segments:
+        f.write(segment.text + "\n")
+print(f"[INFO] Transcription saved to {output_file}")
+print("\n -- Scripted by Syed Sultan (github = Syed-56)")