I timed 4 Mac dictation apps, from letting go of the key to text appearing.
Gabsy: 178 ms
FluidVoice: 343 ms
Wispr Flow: 382 ms
MacWhisper: 676 ms
16 clips, one Mac. FluidVoice is excellent and free, but Gabsy is about 2× faster. Full write up soon.
@PranavOriginals Im building https://t.co/3Bf6tcvy8C
Dictation is just the start. Gabsy also:
• edits what you wrote by voice
• takes meeting notes, offline
• runs Mac commands by voice
• knows your code’s variable names
• works on Mac, iPhone, Android & Windows
Waitlist open 👆
Was not expecting so many signups after so few posts about https://t.co/RQJTouB2Fo so I did not even have a Thank You email set up! Expect one in the next 15 minutes!
@spacedevin Im building https://t.co/3Bf6tcvy8C
Dictation is just the start. Gabsy also:
• edits what you wrote by voice
• takes meeting notes, offline
• runs Mac commands by voice
• knows your code’s variable names
• works on Mac, iPhone, Android & Windows
Waitlist open 👆
@FaisalAbdin90 Plenty of wheels out there. I'm not planning on copying anyone. I have unique features planned that other apps dont. By your logic all dictation apps "reinvented" all the other dictation apps so what is your point?
I'm going to build a Wispr Flow replacement that works fully offline.
Why? My company wouldn't approve Wispr because it sends data off the device. Fair enough. So I'll build one that doesn't.
Building it in public over the next 2 weeks. Follow along 👇
Join the waitlist and get in first 👉 https://t.co/xQovgyAJ06
Today I'm planning the build and writing the feature list of what I need to be a real competitor, not a toy.
Went through Wispr Flow and Superwhisper feature by feature. The list is long. Good.
Waitlist is open: https://t.co/vreqKobvcA. Early access goes there first.
Dictation apps are amazing until IT asks one question: "where does the audio go?"
For a lot of them the answer is "our servers". That's a no at a lot of companies.
So the rule for this build: audio never leaves the device. Ever.
Would you use this? Reply and tell me why (or why not).
Love what you've built, I’ll try it out tonight!!from what I can see this is what Gabsy adds that I can’t see on your site.
• Voice edits by paragraph, with undo • Snippets: say 2 words, get a whole block • A different tone for each app • Literal mode, with no AI rewrite • Voice memo import, transcribed offline • Recovers failed dictations • Zero telemetry
Give me a few weeks and let’s do a speed test :)
@BobzHolmes@FluidVoiceApp I use handy daily! It doesn’t do any of the things that wispr does? It’s just a basic voice to text. It doesn’t clean up your speech or remove filler words! It’s extremely basic but excellent.
Wispr Flow is great. Superwhisper is great.
But Wispr needs the internet to transcribe, and my company won't approve anything that sends audio off the laptop.
So I'm building the one I'm allowed to use. In public. Starting today.
Follow along, I'm shipping a feature every day for 14 days.
You check Search Console and see a search people are clearly typing, and your site has no page for it. Now what? I just shipped this in UpSearch: it finds Search Console queries with no matching page and gives you a page to create for each one.
UpSearch's monthly SEO report now goes out on the first Monday of the month. A report that lands on a random day gets read on a random day. This way it's a fixed point you can plan around. Same plain answer: what changed, the pages and searches behind it, and the next step.
Product question for anyone who has shipped SEO software. When a tool lacks data to answer, I make UpSearch say "not enough evidence" instead of giving a best guess. It's less satisfying to read. Do users trust that, or do they just want an answer?