← All patterns
Voice & Realtime AI
Conversation under a latency budget: turn-taking, barge-in, and holding the floor.
5 patterns
Cascaded vs. Speech-to-Speech
The first architectural decision in any voice product: three composable models in a chain, or one model that hears and speaks directly.
9 minaka STT-LLM-TTS pipeline
Turn-Taking and Barge-In
Decide when the user has finished speaking and what happens when they talk over you — the two mechanics that make a voice agent feel human or broken.
10 minaka endpointing
Latency Budget to First Audio
Treat time-to-first-audible-syllable as the product metric, and engineer every stage against an explicit allocation.
9 minaka time to first byte of speech
Async Tools While Holding the Floor
Keep talking while a slow tool runs in the background, and weave its result in when it lands.
9 minaka backgrounded tool calls
Readback Confirmation
Read critical values back in a disambiguated form and get explicit confirmation before acting on anything transcribed.
9 minaka voice grounding