On-device voice isolation, wake words, voice activity and voiceprints. In the browser (WebAssembly) and in Python, with plugins for LiveKit Agents and Pipecat. Apache-2.0 code and weights.
Turn your own spoken words into a finished article. AI does the research, structure and checking. You supply the sentences, by talking.