Recent Posts

A man examines a severed cord on an early candlestick telephone.

voice-ai

Are There Enough Phone Calls to Go Around for Hundreds of New Voice AI Vendors?

Calling minutes are not disappearing. European usage has shifted toward online platforms, while India’s regulated calling market continues to grow.

Illustrated 1926 magazine page explaining how photographs were scanned, transmitted across the Atlantic by cable, and reconstructed at the receiving station.

voice-ai

AI Agents Can Build Apps. Why Can’t They Use Them?

A few weeks ago I rage-coded a fax app (promptfax.app) after I was forced to send a fax again for the first time in a decade. I did not want to deal with sending a one-off fax, so like I often do with stasks I don’t

voice-ai

Voice AI, WebRTC, and MoQ

Some initial commentary on whether Voice AI needs all of WebRTC and where Media over QUIC or machine-oriented media formats could fit.

voice-ai

Inside OpenAI’s gpt-realtime WebRTC Implementation

A condensed look at the WebRTC negotiation, ICE changes, video path, and ChatGPT comparison observed in OpenAI’s gpt-realtime release.

voice-ai

How OpenAI Built Its Realtime WebRTC Gateway

Highlights from my discussion with Sean DuBois about OpenAI’s WebRTC gateway, WebSockets, ICE, audio, scaling, SIP, and embedded devices.

voice-ai

OpenAI’s Realtime WebRTC API: A Guide to the Beta Implementation

A condensed guide to the architecture, media flow, DataChannel events, function calling, and AIY Voice Kit experiment in OpenAI’s Beta Realtime WebRTC API.