Archive - Week in Voice AI

#31: Wispr Flow to overhaul quality with new models

Investors are pouring money into speech generation startups

Aug 3 • Ivan Mehta

July 2026

Wispr Flow is building a meeting notetaker

The company's changed terms hint towards a new feature

Jul 31 • Ivan Mehta

#30: ChatGPT and Claude voice get promising, but clunky

Voice AI assistants are good, but not good enough for a workout

Jul 27 • Ivan Mehta

#29: ChatGPT, turn on my light

OpenAI is building a smart speaker

Jul 20 • Ivan Mehta

Hume AI's life after the Google deal

And its new Voice EQ benchmark

Jul 17 • Ivan Mehta

#28: Chatting with ChatGPT

OpenAI's big bet on voice models

Jul 13 • Ivan Mehta

#27: Local processing (Acoustic version)

The post-tokenmaxing era: Local AI via sound waves

Jul 6 • Ivan Mehta

June 2026

#26: Heidi's health hardware

Why is it important to own voice stack in the health world

Jun 29 • Ivan Mehta

Why an Icelandic company built a new ASR benchmark with Hugging Face

The benchmark measures how models fare in real-world scenarios

Jun 25 • Ivan Mehta

#25: India's biggest telco gets in on voice play

Vapi's fun experiment lets you figure out how 'Human' is an AI voice

Jun 22 • Ivan Mehta

#24: Apple gets Siri-ous about its assistant

Jun 15 • Ivan Mehta

#23: Lost in translation

Companies need to work on better translation tools in the era of large language model

Jun 8 • Ivan Mehta

#nojs-banner { position: fixed; bottom: 0; left: 0; padding: 16px 16px 16px 32px; width: 100%; box-sizing: border-box; background: red; color: white; font-family: -apple-system, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol"; font-size: 13px; line-height: 13px; } #nojs-banner a { color: inherit; text-decoration: underline; } This site requires JavaScript to run correctly. Please turn on JavaScript or unblock scripts