A lightweight, CPU-only audio transcription powered by faster-whisper. Utilizes the efficient tiny.en model to deliver fast and accurate transcription. Supports output in plain text or SRT subtitle format.