A speech recognition service that converts audio input into text. It supports real-time and batch transcription, custom vocabulary, and multiple languages.