Speech Note is a desktop and mobile application for note taking that leverages multiple offline ML models for audio and text processing. It performs speech-to-text transcription using engines like whisper.cpp, Faster Whisper, Vosk, and Coqui STT, synthesizes speech using Piper, Coqui TTS, Mimic 3, and other TTS engines, and translates text between languages using neural machine translation models. All processing runs locally on-device with no network connectivity required.