Everything on voice and chat AI agents — real-time speech processing, telephony and phone-system automation, conversational AI, voice UX design, and production deployment.
From the blog
Spanglish, Hinglish, and Arabic-English breaks most voice stacks. Deepgram's code-switching voices, LiveKit auto-language detection, and a CallSphere bilingual flow that holds context across language flips.
Tavus Phoenix-4 hits sub-600ms end-to-end. Here is how 2026 chat agents return short avatar videos, switch to live video calls, and bill at $0.23–$1.09 per minute.
VAPI deprecated 'Custom Functions' for the Tools API. Wire a webhook tool, secure with HMAC, and ship a working appointment-booker — code + pitfalls.
Mint an ephemeral OpenAI key from a Next.js Route Handler, connect via WebRTC from the browser, and ship a working voice demo to Vercel in one afternoon.
Where TCPA $500-$1,500 per-call exposure actually comes from, how AI voice changes the consent calculus, and the operational controls that keep you under the line in 2026.
Build a working voice agent with Coqui XTTS-v2: clone any voice from 6 seconds of audio, speak 17 languages, run it locally with no API keys. Full setup.
Stand up a Gemini-powered voice agent with Vertex AI Agent Builder (now Gemini Enterprise Agent Platform). Phone gateway, ADK code-first agent, Cloud Run runtime — under 200 lines.
Live faith services in 2026 ship multilingual AI captions over WebRTC to congregations spanning 100+ languages. Here is the production stack with on-prem ASR, accessible overlays, and donation flows.
Add an AI agent to a Twilio Conference as a first-class participant via a TwiML Application. We cover the Add Participant API, mute/coach roles, and CallSphere's three-way escalation pattern.
