The sound of typing has been the background hum of office work for a century and half. But now it’s all about whispers.
After years of bashing typewriters, then tapping keyboards, desk-bound employees are, in ever-increasing numbers, murmuring to AI dictation apps to send emails, draft reports and write code.
‘Double words per minute’
Voice-to-text software has been around since the 1960s but it was always “clunky” and slow and “never worked quite right”, said employment reporter Jo Constantz on Bloomberg. But now advances in AI have made it “viable”: it can “take the messiness of speech and package it into something more useful”.
The Week
Escape your echo chamber. Get the facts behind the news, plus analysis from multiple perspectives.
SUBSCRIBE & SAVE
Sign up for The Week's Free Newsletters
Join more than 350,000 subscribers and keep yourself informed with a selection of The Week’s most interesting, enlightening and entertaining stories - plus daily puzzles.
Join more than 350,000 subscribers and keep yourself informed with a selection of The Week’s most interesting, enlightening and entertaining stories - plus daily puzzles.
Latest Videos FromThe Week
Early adopters of AI dictation apps are “drawn inexorably to the promise of ever-greater productivity”. In “voice mode”, you can produce double the words-per-minute than you can when typing.
Dictation is definitely “having a moment”, said Joe Castaldo, business reporter at Canada’s The Globe and Mail. More and more software engineers, in particular, are switching from “pressing keys individually” to “adopting AI-powered speech-to-text apps to verbally issue instructions” to tools such as Anthropic’s Claude Code. Eight months ago, internet entrepreneur Reid Hoffman posted on his LinkedIn platform that he has been “voicepilled”: he’d realised you can “amplify your ability” by “seriously using your voice to interact with technology”.
Start-ups today are like “a high-end call centre – except everyone is chatting with AI”, one venture capitalist told The Wall Street Journal. There is an “etiquette”: “users try to keep their voices low and often wear headphones to block out sound from their dictating neighbours, dialling down the annoyance factor”. But talking to yourself is still “weird, if not a little embarrassing”.
‘Velocity towards voice’
It’s too early to say if and when “the Qwerty keyboard might follow the ticker tape and fax machines into obsolescence” but “the velocity towards voice is accelerating”, Dylan Fox, CEO of San Francisco-based AssemblyAI, told the Los Angeles Times. We’re predicting a 10 to 100-fold “increase in demand for voice, AI applications and interfaces”.
There’s now “a mad dash to dominate any corner of the evolving field”, said Bloomberg’s Constantz. The market for AI voice generators alone is estimated to be worth $7.7 billion (£5.75 billion) this year, rising to $21.8 billion (£16.27 billion) by the end of the decade, according to US consulting firm Grand View Research.
Google, Apple and Microsoft have invested heavily in their voice-to-text products, and dictation app start-ups – many with variations of “whisper” in their name – have experienced remarkable growth over the past year. After all, Superwhisper founder Neil Chudleigh told The Globe and Mail, “we’re talking about replacing every keyboard on the planet”.
