Your keyboard is the slowest part of your working day and you have no idea what it costs you.
Take a look at my own numbers from the last three months: 167,820 words dictated! But what is the impact? An average person types around 45 words a minute and speaks around 150. So an hour of typing produces about 2,700 words, and an hour of talking produces about 9,000. Same hour, three times the output. I spend the difference on longer prompts and fuller answers to my team, and I didn't add another hour to my day to get there :)
My own numbers are the small story. Worldwide, people have run 683,838 phrases through Lispr and skipped about seven months of typing. Seven months! Counted in working hours that is 1,232 of them, one person sitting at a keyboard eight hours a day from January through to the end of July, and nobody had to be that person. Those hours went straight back into people's lives.
My share of those hours came from three things, and copying one is the quickest way for you to start saving your time.
I talk to Claude Code and Cursor instead of typing at them. I hold the key and describe the problem: what the module does, what I already tried, which constraint matters, where the data comes from. The text appears in the prompt box when I release the key. Typing all that, I would have trimmed it to three lines to spare my fingers, then spent two more rounds explaining what I left out. Speaking costs me nothing, so the details stay in, and the first answer needs less fixing.
I dictate in my native language and the text arrives in English. When you are tired, speaking a second language costs more effort than the work itself. You can say it in your own language instead, hold one extra key, and the reader gets English or any of 32 languages that Lispr translates to.
I taught it the words that matter to us. Every project has terms no transcription model knows: client names, service names, internal shorthand. I added ours to the vocabulary list once, and they come out spelled correctly now. The app looks at recent dictations and adds new ones it has seen me fix. People keep asking us for this, so it is worth saying plainly that it already ships, it sits in the settings panel, and it is the feature I would miss first.
One question for everyone using it. What does your voice-to-AI workflow look like?
Replies
167k words is wild 😂 At that point your keyboard deserves a vacation. I can definitely see why longer prompts become easier when talking feels almost effortless.
For me, it is also prompting! I thought that touch-typing was a game-changer for me, but I was wrong) Now, I can't imagine using Claude without Lispr, because it's not only faster, but also much easier (not moving your fingers across the whole keyboard..)
Lispr
Adding to Konstantin's third point, because the vocabulary list is the part most people never find.
It sits in Dictionary in the sidebar. You give it two things: how a word should be written, and what Lispr hears instead. The list lives on your device.
There is also a switch for Auto-expand. With it on, Lispr goes through your recent dictations and picks out the names and terms it has been getting wrong, so the list grows without you maintaining it. There is a button to run that pass on demand.
For technical vocabulary and names, it makes a real difference for me.