Realtime Voice Agents

The chapters in this module, in reading order.

#	Chapter
00	Voice & Realtime AI — The Five-Year-Old Version
01	Five-Second Failure — Slow sounds stupid
02	Audio Processing Basics — Learn the raw signal first
03	Streaming ASR — Teach the ear to listen live
04	Text-to-Speech — Make the voice arrive early
05	Streaming Pipeline — the relay race must keep moving
06	Latency Budgeting — name every millisecond before it embarrasses you
07	Interruption and Barge-In — when the user talks over the assistant
08	End-to-End Voice Models — fewer handoffs, different trade-offs
09	Telephony constraints — phone calls humble shiny demos
10	Evaluation and debugging — voice systems punish vibes
11	Honest admission — voice and realtime AI still fight physics