🚀 Python Desktop Tool – Mp3toText

The Mp3toText tool automatically processes Vietnamese audio files and converts them into text.
With a very simple interface, just a few clicks allow users to turn full conversations into text for practical use.

---

✨ Key Features

• 📝 Choose between Whisper (OpenAI) or PhoWhisper (VinAI) models, downloaded locally.
• ⚡ Automatically transcribe audio into text before feeding it into any system.
• 🔄 Speaker gender detection (male vs female) using clustering based on voice frequency (Hz).
• 🔄 Export results with a clear and intuitive display.
• 🖥️ User-friendly interface, no programming skills required.
• 🛠️ Compare models: users can evaluate whether OpenAI’s Whisper or VinAI’s PhoWhisper transcribes Vietnamese more accurately.

---

📖 Approach

Whisper Model – OpenAI

PhoWhisper Model – VinAI

📄 View PhoWhisper Paper (VinAI)

---

📸 Application Interface

Main screen on launch

Processing data with models

Final transcription results

---

🛠️ How to Use

1. Prepare your audio file in formats like .mp3 or .mp4 (example: fileAudio.mp3).
2. Open the application Mp3toText.exe.
3. Select the audio file to process.
4. Click Run to start automatic transcription.
5. View results directly on the interface or from the output file.

---

📦 Installation

1. Clone the repository:

   git clone https://github.com/Phuc75nguyen/MP3toText.git
   cd AutoData4FA

2. Set up a virtual environment:

   python -m venv venv
   venv\Scripts\activate

3. Build with PyInstaller in VS Code:

   python -m PyInstaller --noconfirm --onefile --windowed --name MP3toText --icon=app.ico --add-data "app.ico;." app.py

4. Run the app:

   python app.py
   

❤️ Made with Tons of Love (Tan Phuc)