Runtime Speech Recognizer

High-performance OpenAI's Whisper speech recognition

Explore the docs »
Fab . Discord support chat

Key features

• Fast recognition speed
• English-only and multilingual models available, with multilingual supporting 100 languages
• Different model sizes available (from 75 Mb to 2.9 Gb)
• Automatic download of language models in the Editor
• Optional translation of recognized speech to English
• Customizable properties
• Easy selection of model size and language in settings
• No static libraries or external dependencies
• Cross-platform compatibility (Windows, Mac, Linux, Android, iOS, etc)

Additional information

The implementation is based on whisper.cpp.

Like my work?

Consider supporting me. Hire me at [email protected].