Wispr Flow Has Transformed Modern Communication

Voice notes are a gift to the sender and a chore for the receiver. It’s far more pleasant and convenient to simply speak out your thoughts than type them out — particularly for a long message — but it takes so much longer to listen to a message than to read one. People also talk far less concisely than they write, so a message that would otherwise be communicated in a few crisp sentences tends to take half a minute or more to say.It’s with this in mind that a range of startups have popped up promising to use AI voice-dictation models to replace your keyboard with your microphone. The market leader is Wispr Flow, which I’ve been using for more than a year now.Hit the shortcut button — holding down for quick messages, or pressing a different one to engage the “hands-free” mode for longer messages — and whether you’re writing code, sending an email, or narrating the first draft of an article, you can just start talking. A short beat after you release the shortcut — or hit the hands-free shortcut again — your text box will fill with nicely formatted text. You can use it on iOS, Android, Windows, and Mac, and though all of these have alternative voice-typing modes — such as the native voice keyboard on iOS — the advantage with tools like Wispr is far more reliable transcription, a customizable dictionary and replacement shortcuts, and dynamic, intelligent formatting.If you’re putting down thoughts for a memo, it will simply be a single block paragraph; but if you are instead sending an email, it will space it out and format it appropriately. So long as you keep it in the “Formal” setting, it will also format your text with correct grammar.Wispr has other tweaks, too. Snippets let you insert longer text extracts — you can say “my address is” and have a snippet saved to autofill the details — and you can add custom words and replacements to the app’s dictionary. And if you want even more complex formatting — turning your messages into well-formatted AI prompts, or just dramatically condensing them — you can set these up with the new “Transforms” tab, where you speak out your text and then hit a keyboard shortcut.This is not to say it’s perfect, though. For one thing, Windows consistently has a short delay between when you press the activation button and when it actually starts recording you. You can reduce this by setting a single microphone as the default rather than using the system’s auto-detect, but it’s still not instant; it takes a while to learn that you have to wait a beat before you start talking. When I first started using it, I would talk before it started recording, only to paste text missing the first three or four words.It’s also not the only option, and its main competitor, Aqua, has a few features that I preferred:The transcription is just as accurate but much faster. Wispr leaves only a small pause after you finish speaking before the text appears, but Aqua has almost none — and that speed makes you want to use it more often, particularly for short messages.It has better context awareness: so long as you allow it, it reads what’s on your screen and fills in the details based on that context.In hands-free mode, Aqua has a desktop setting that shows the text you have already said as you continue speaking, which is both pleasant and helpful.Aqua has custom, app-specific preferences that you can write in natural language and that allow more customization than Wispr. For example, when I’m messaging my friends, I use proper grammar and capitalization but far more exclamation marks than I do in my writing — and you can’t specify that on Wispr. This sounds like a small thing — and it is — but the more you have to edit the transcribed text, the less you want to use the tool.For a few months, Aqua had won me over, but it has proven both buggier and less adaptable to Apple’s changes than Wispr. Aqua transcribes faster, but for the past month, at least 50 percent of the time when I hit the button to finish a recording on iPhone, it doesn’t paste the text and instead displays an error message. It has transcribed the message correctly, but to insert it, I have to tap the three dots on the right, hit the “Recover Text” button, and then it will paste. It’s not terribly inconvenient, but it’s also unnecessary, and having to click two buttons for every transcript completely undoes any speed benefit it has over Wispr.The other issue is that Apple has cracked down on permission settings for voice keyboards with recent iOS changes, and Wispr has been better — in my testing — at circumventing Apple’s finicky restrictions. Whenever I try to send a message on Aqua, even if it’s in the same app, the keyboard switches to the Aqua app, which I then have to swipe back from. With Wispr, I occasionally need to do this, but its Live Activity — and its Control Center microphone toggle — have done a better job of keeping the Apple restrictions at bay.Wispr needs to improve its latency, add more customization settings, and probably offer a local transcription option, given how good small offline AI models are getting. But despite new rivals coming on the market and a year of AI development, Wispr is, in my experience, still the best option around for voice dictation, particularly if you use multiple platforms.You can use it for free to send up to 2,000 words a week on desktop, or 1,000 on iPhone. Pro gets you unlimited messages at $12 a month on the annual plan — which I just renewed — or $15 monthly.

Wispr Flow Has Transformed Modern Communication

More from the New York Sun

Back to the USSR, Russian Wait in Gas Lines — and Fume

Debbie Wasserman Schultz and the Democrats’ New Voting Math

Trump-Backed Candidate for Georgia Governor Loses His Primary to Billionaire First-Time Candidate

More from the New York Sun

Back to the USSR, Russian Wait in Gas Lines — and Fume

Debbie Wasserman Schultz and the Democrats’ New Voting Math

Trump-Backed Candidate for Georgia Governor Loses His Primary to Billionaire First-Time Candidate

Democrats Withhold Judgment on Clayton’s Nomination To Be Spy Chief, Choosing To Wait for Answers at Confirmation Hearing

Warsh’s First Press Conference as Fed Chairman Could Offer Clues About His Promised Reforms

Justice Department Seeks To Join Lawsuit Challenging Evanston’s Reparations Program

Hegseth Presses GOP Lawmakers for $350 Billion Pentagon Spending Increase Atop $1.1 Trillion Defense Appropriations

After Court Loss, East Village Group Says Fight Over Mamdani’s Homeless Intake Center Plan Is Far From Over