HOOK
When did you last write something true?
Not an email. Not a report. Something true. An idea that mattered to you, expressed without filter, without self-censorship, without that inner gaze that judges every word before it lands on the screen.
For me, it was three weeks ago. I had an idea for a PerkySue feature. I sat down in front of my editor. The cursor blinked. And for twenty minutes, I wrote nothing. Not because I had no ideas. Because the keyboard forced me to judge before creating.
Blank page syndrome isn't a lack of inspiration. It's a dysfunction between thought and expression, created by the interface itself. The keyboard, that wonderful tool, is also a psychological barrier between what you think and what you write.
And what if the solution wasn't to "work harder" or "get more organized," but simply to change the interface?
PART 2 — SYSTEMIC PROBLEM
But the problem runs deeper than my personal frustration.
Speaking is natural. We evolved to speak. Speech is our primary mode of expression, developed over hundreds of thousands of years. Writing, by comparison, is a recent artifact — barely a few thousand years old. And the keyboard? An artifact of an artifact, barely a century old.
When you speak, ideas flow. Transitions happen naturally. Prosody carries meaning. Hesitations are normal, productive, creative. When you type on a keyboard, every word is a conscious decision. Every sentence is a construction. Every paragraph is scaffolding. The brain switches from "fluid creation" mode to "permanent editing" mode — and these two modes are neurologically incompatible.
The result is chronic performance anxiety. The blinking cursor becomes a judge. Every typed word is submitted to an inner tribunal: is it good enough? Is it clear enough? Is it professional enough? This real-time self-censorship kills creativity. You cannot create and judge simultaneously. The keyboard forces you to.
And then there's fatigue. Typing is a repetitive physical act. Fingers tire. Wrists ache. Shoulders tense. Concentration fragments between content and mechanics. You're thinking about both WHAT you want to say and HOW you type it. It's exhausting cognitive multitasking.
But the most insidious problem is surveillance. When you know that every keystroke is potentially recorded, analyzed, used to train a model — even unconsciously — you self-censor more. The keyboard isn't neutral. It's a control tool disguised as a creation tool.
PART 3 — THE ALTERNATIVE
Voice dictation removes this barrier. It lets us return to the mode of expression we were made for: speech. You think, you speak, words appear. No mental translation. No real-time judgment. No physical fatigue. Just the natural flow from thought to expression.
The mechanism is simple but profound. When you speak, you can't self-correct every word. You have to hold the thread to the end. You have to accept temporary imperfection to reach global perfection. It's pure creative mode — the one the keyboard prevents.
But cloud dictation adds an additional psychological problem: surveillance. Knowing your voice is sent to a remote server, analyzed, potentially stored, adds a layer of anxiety. You hesitate to say certain things. You self-censor. You lose the spontaneity that makes dictation powerful.
Local artificial intelligence — models that run entirely on your machine, with no internet connection, no account, no subscription — solves this dilemma. It offers the fluidity of speech without the barrier of surveillance.
OpenWhispr (macOS, Windows, Linux, MIT, free) offers local transcription with Whisper, with an optional BYOK cloud mode. Handy (macOS, Windows, Linux, MIT, free) is the open-source standard with 22,435 stars and a very active community. VoiceInk (macOS only, GPL v3.0, $25-49 one-time or free build) is the Apple reference with native AI modes and a personal dictionary. And PerkySue (Windows, Apache 2.0, free for transcription, $9.90/month Pro) adds a local LLM transformation layer — rewriting, email, console, AI answers — all without your voice leaving the machine.
None of these tools is perfect. OpenWhispr still requires some technical setup. Handy doesn't have integrated AI modes as advanced as commercial tools. VoiceInk is Mac-only. And PerkySue — my tool — is Windows-only, with no macOS version. Each has its limits.
But what they offer is a private space where thought can flow without filter. Your hesitations stay private. Your false starts are never judged. Your voice never leaves your computer. This privacy isn't a luxury. It's a condition of creativity.
PART 4 — INDIVIDUAL IMPACT
Local dictation transforms the relationship with writing for everyone who suffers from blank page syndrome.
You can dictate an entire draft in minutes, without ever seeing the cursor blink. You can rephrase an idea by speaking, without going through the keyboard. You can generate content at the speed of thought, then switch to editing mode to polish.
But beyond productivity, there's mental well-being. No more cursor anxiety. No more real-time judgment. No more physical fatigue. Dictation lets you separate creation from editing — two distinct phases, each in its optimal mode. Create by voice, edit by keyboard. It's the method that respects the brain.
And there's autonomy. Local AI works everywhere — on a plane, in a noisy café, in a coworking space. No connection, no network dependency. You can dictate your ideas anywhere, anytime, without technical or psychological barriers.
I'm not saying local dictation will cure performance anxiety. But I know that when I see the cursor blinking and my brain freezes, pressing Alt+T and speaking changes everything. It's not magic. It's just neurologically more honest.
PART 5 — COLLECTIVE STAKE
The stakes go beyond the individual. When teams adopt local dictation tools, they unlock collective creativity.
A company whose employees can express ideas at the speed of thought produces more, better, faster. A team that dictates its meetings captures collective intelligence without loss. An organization that removes the keyboard barrier democratizes contribution.
Open source is the foundation of this liberation. Accessible, modifiable, auditable source code guarantees that the tool will evolve with creative needs. That a community of developers can add modes adapted to different types of creation. That no company can unilaterally impose a format or method.
PerkySue, under the Apache 2.0 license, embodies this philosophy. Every line of code is visible. Every user can verify what the tool does — and ensure it doesn't judge their ideas, correct their personality, or standardize their expression. A competent developer can modify the code, add creative modes, create a version adapted to specific needs. The tool belongs to its community.
Adopting local dictation isn't rejecting technology. It's a choice of creative intentionality. It's consciously deciding which tools deserve our trust, which barriers we want to eliminate, which forms of expression we want to liberate.
It's also a question of real cost. Cloud subscriptions accumulate — dictation, storage, office tools, creation. For a freelancer, a student, an independent creator, these costs are prohibitive. Local dictation offers an alternative where the cost is transparent, controlled, and above all — fair. No financial barrier, no technical barrier, no psychological barrier.
Next time the cursor blinks and your brain freezes, ask yourself: do I lack ideas, or does the tool prevent me from expressing them? If the answer liberates you, maybe it's time to speak up.
CONCLUSION
We've built tools that force us to judge before creating. That make us translate our thoughts into a language the machine understands. That exhaust us because they demand we do two things at once — think AND type, create AND edit, express AND censor.
Maybe it's time to accept a simple truth: our brains didn't evolve to type. They evolved to speak. And the best writing tool might not be a keyboard. It's our own voice.
Are you ready to trust it?
About the Author
Jérôme Corbiau is the creator of PerkySue, a local voice dictation tool with AI that works entirely offline, with no remote server or data transmitted. He is also co-founder and software architect of My App Zone SRL (Brussels), and creator of the Cloud Neareo platform — an award-winning CMS notably by Microsoft and the Public Service of Wallonia, deployed in museums and heritage sites. His work aims at a constant objective: putting technology at the service of the user, rather than the reverse.
P.S. — If local voice dictation interests you, I've open-sourced what I use daily: github.com/PerkySue/PerkySue. No account, no cloud, just a hotkey. Windows only for now — and I know that's a limitation.
Top comments (0)