Seda2Text 🎙️

Seda2Text is a desktop application that converts Persian (Farsi) speech from audio or video files into text subtitles (.srt). It uses the powerful OpenAI Whisper model via the Hugging Face Transformers library and provides a clean graphical interface built with PyQt6.

🧩 Features

-Beautiful, user-friendly GUI built with PyQt6

-Supports audio and video input formats (.mp3, .wav, .mp4, .mkv, etc.)

-Uses the Persian Whisper model (vhdm/whisper-large-fa-v1) for accurate transcription

-Option to use a local Whisper model for offline transcription

-CUDA GPU acceleration for faster processing (if available)

-Generates synchronized SRT subtitle files

🧠 Requirements

-Python 3.9 or later

-FFmpeg installed and added to your system PATH

-(Optional) NVIDIA GPU with CUDA support

📦 Installation

1- Clone the repository:

git clone https://github.com/yourusername/seda2text.git
cd seda2text

2- Install dependencies:

pip install -r requirements.txt

3- Check FFmpeg installation:

ffmpeg -version

🚀 Usage

1- Run the application:

python main.py

2- In the GUI:

Select a video or audio file.

(Optional) Check “Use a local Whisper model” and choose your model directory.

Click Generate Subtitles. The generated .srt subtitle file will be saved in the same directory as your input file.