Video Subtitler & Text-to-Speech Generator

Overview

This project provides two standalone tools for Video Subtitling and Text-to-Speech generation using OpenAI's models. The Video Subtitler extracts and transcribes audio from video files, while the Text-to-Speech Generator converts text files into speech. Both tools are distributed as .exe files for easy use, and they leverage OpenAI’s Whisper for transcription and TTS for generating speech.

Features

Video Subtitler

• Transcribes audio and video files using OpenAI's Whisper speech-to-text model.
• Supports video formats like .mp4, .mov, .mpeg, and more.
• Extracts audio from video files and splits large audio files for optimal transcription.
• Offers transcription correction using GPT for refined and accurate text output.
• Customizable with language and prompt settings.

Text-to-Speech Generator

• Converts text files into speech using OpenAI's TTS model.
• Customizable voice selection and speed settings for more control over the output.
• Outputs audio as .mp3 files.

Prerequisites

FFmpeg Installation

FFmpeg is required to handle video and audio processing. Install FFmpeg based on your operating system:

• Windows Option 1: Install using winget (recommended):

  1. Open Command Prompt as Administrator and run:

     winget install ffmpeg

• Windows Option 2: Download from the official website:

  1. Download the latest build from the official website.
  2. Extract the ZIP file to a location on your computer, e.g., C:\ffmpeg.
  3. Add the bin directory of ffmpeg to your system's PATH.

• macOS: Install via Homebrew:

  brew install ffmpeg

• Linux: Install via APT:

  sudo apt-get install ffmpeg

OpenAI API Key

• Obtain an OpenAI API key from OpenAI's platform.
• Set this API key in the config.yaml file as described in the configuration section.

Installation

Windows

1. Download the ZIP:

   • Go to the Releases page on GitHub and download the latest .zip file.
   • Extract the contents to a directory on your system.

2. Set Up Configuration:

   • Rename the config.example.yaml file to config.yaml and set up the configuration parameters.
   • Basically you only need to set the openai.api_key parameter with your OpenAI API key.
   • The prompt parameter can be used to teach Video Subtitler about special words or phrases (trademarks, unusual names) that may appear in your video.
   • Example config.yaml:

     openai:
       api_key: sk-XXX
       stt_model: whisper-1
       tts_model: tts-1
       completions_model: gpt-4o
       temperature: 0
     default:
       language: EN
       stt_prompt: PhraseVault, Video Subtitler
       tts_voice: echo
       tts_speed: 1

macOS/Linux

1. Clone the Repository:

   • Clone the repository to your local machine:

     git clone https://github.com/ptmrio/video-subtitler.git
     cd video-subtitler

2. Set Up Configuration:

   • see step 2 in the Windows installation instructions.

3. Install Dependencies:

   • Install the required Python packages:

     pip install -r requirements.txt

4. Run the Application:

   • Run the application using Python:

     python video_subtitler.py

     or

     python text_to_speech.py

Usage

Text-to-Speech Generator

1. Navigate to the downloaded and extracted folder and run the text-to-speech.exe file.
2. Enter the path to your .txt text-file, customize the voice and speed (if necessary), and click Generate Speech.
3. The application will convert the text into speech and save the result as an .tts.mp3 file.

Video Subtitler

1. Navigate to the downloaded and extracted folder and run the video-subtitler.exe file.
2. Provide the path to your audio or video file and configure optional settings such as language or custom prompts.
3. Click Transcribe to begin transcription. The resulting transcription will be saved as a .transcription.txt file.

License

This project is licensed under the MIT License. See the LICENSE file for more details.

Donations

If you find this project useful, consider donating to support its development.

Thank you for your support!