This path is for you if voice is a speak button and you need the split underneath: text to speech, speech to text, a cloned voice, and the delay that makes a conversation feel broken. No part is live on this page yet. Start with the product series for the voice button you already have, and come back when part 1 is up.
Foundations & Acoustic Architecture
- 1 TTS, STT, and Voice Cloning: The Three Mental Models of Speech AI Separate TTS, STT, and cloning mentally. Each has different quality knobs, consent rules, and failure modes. Match the tool to the audio task. Scheduled · January 8, 2027