Subtitle Extractor

Subtitle Extractor

Turn videos and audio files into subtitle files using fast and fully local AI processing.

Subtitle Extractor analyzes the audio from your selected file, converts speech into text, and exports the result as a subtitle file. The AI models run directly on your computer, so your videos and audio files do not need to be uploaded to an external service.

For speech recognition, the application uses Whisper Large V3 Turbo Q8 to provide accurate transcription. It also includes three GemmaTranslate model options, allowing you to choose between lower hardware usage, balanced performance, or higher translation quality.

Main features

• Automatically extract subtitles from video audio.
• Transcribe standalone audio files.
• Generate subtitles in Portuguese and English.
• Speech recognition powered by Whisper Large V3 Turbo Q8.
• Three GemmaTranslate translation levels:
  • Lightweight: lower memory and processing requirements.
  • Balanced: a balance between speed and translation quality.
  • High Quality: improved translation accuracy and fidelity.
• Fully local processing.
• GPU acceleration for supported models.
• Built-in video preview.
• Transcription and translation progress indicators.
• Automatic model download and management.

Supported formats

Subtitle Extractor supports several common video and audio formats, including:

Video: MP4, MKV, MOV, AVI, and WebM.

Audio: MP3, WAV, M4A, FLAC, OGG, and AAC.

Available models

The application uses four main AI models:

• Whisper Large V3 Turbo Q8 for speech recognition.
• GemmaTranslate Lightweight for lower-end hardware.
• GemmaTranslate Balanced for general use.
• GemmaTranslate High Quality for improved translation fidelity.