Why Modern Creator?

Paul J Lipsky · YouTube

Wispr Flow Got Me Addicted to Voice Prompting

A sponsored walkthrough of how talking instead of typing changes what you actually say to Gmail, Gemini, and ChatGPT.

Posted

July 7th

1 months ago

Duration

07:29

Format

Demo

educational

Views

10.2K

331 likes

Big Idea

The argument in one line.

Voice-to-text tools don't just transcribe faster than typing -- they let you produce cleaner, more context-rich AI prompts by talking naturally, and that combination is what actually changes a daily workflow.

Who This Is For

Read if. Skip if.

READ IF YOU ARE…

You regularly prompt AI tools like ChatGPT or Gemini and want faster, more detailed inputs without typing everything out.
You write a lot of emails or messages and want dictation that also formats and cleans up the text automatically.
You're a mobile user who avoids typing long messages or notes on your phone because it's slow or error-prone.

SKIP IF…

You already have a dictation workflow you're happy with and aren't looking to switch tools.
You want an independent technical comparison of dictation apps rather than one creator's sponsored use-case walkthrough.

TL;DR

The full version, fast.

Wispr Flow is a voice-to-text tool the creator uses instead of typing for emails, AI prompts, and phone notes. Beyond raw accuracy, it auto-formats output based on context (email vs. prompt), strips filler words and self-corrections, learns custom spellings from edits, and lets users save reusable 'snippets' triggered by a spoken cue. The core case: because talking is faster than typing, people give AI tools more context and more detailed prompts than they would if they had to type it out, which improves the AI's output. The video demos this across Gmail, Gemini, ChatGPT, and the iOS keyboard, positioning the tool as a habit change rather than just a transcription upgrade. It's sponsored content with a Wispr Flow discount code.

Free for members

Chat with this breakdown — free.

Sign in and you get 23 free chat messages on us — ask for the hook, quote a framework, find the exact transcript moment, generate a markdown action plan. Bring your own key when you want unlimited.

Create a free account →

Chapters

Where the time goes.

00:00 – 00:57

01 · Intro

Hook states the thesis (voice-to-text as a prompting upgrade, not just transcription) and discloses the Wispr Flow sponsorship with the free-month link.

00:57 – 01:37

02 · Writing Emails by Voice

Dictates a full email reply in Gmail; Wispr Flow formats it as a proper email, strips filler words, and resolves an in-sentence correction.

01:37 – 02:17

03 · Wispr Flow vs Built-In Dictation

Side-by-side: the same spoken email run through built-in dictation comes out unformatted with every filler word and the failed correction left in.

02:17 – 04:06

04 · Prompting Gemini

Live prompt to build a one-page restaurant website with specific details; catches a misspelled business name and shows Wispr Flow auto-learning the correct spelling.

04:06 – 05:29

05 · ChatGPT and Snippets

Settings, dictionary, and snippets carry over to ChatGPT; a spoken trigger phrase instantly types out a long saved image-cleanup prompt.

05:29 – 07:01

06 · Voice Prompting on Phone

Overhead shots dictating on an iPhone -- a quiet workout question, then a numbered note-taking list, formatted automatically.

07:01 – 07:29

07 · Outro

Recap plus a second CTA with an on-screen promo code overlay for a free month of Wispr Flow Pro.

Atomic Insights

Lines worth screenshotting.

Voice dictation isn't just faster than typing -- the real gain is that people write longer, more detailed AI prompts when they can just talk instead of type.
Wispr Flow formats text differently based on context: the same spoken words come out as a signed email in Gmail but a clean prompt in a chat box.
Correcting a misspelled word once, like a business name, can teach a dictation tool that spelling permanently, without opening any settings menu.
A single spoken trigger phrase (a 'snippet') can insert an entire saved prompt, so a complex, detailed instruction takes the same few seconds as a short one.
Built-in phone dictation often fails badly enough that people give up on tasks like note-taking entirely; a better voice tool can restore habits people had abandoned.
Reusing the same saved prompt via a snippet keeps AI output more consistent than re-typing or re-speaking a similar instruction each time.

Takeaway

How talking to AI beats typing your prompts

WHAT TO LEARN

Voice input's real advantage isn't speed -- it's that people give AI tools more detail and context when they can just talk, and a tool that formats output and remembers corrections turns that habit into consistently better results.

02Writing Emails by Voice

Dictation tools that understand context can auto-format the same spoken words differently for an email versus a chat prompt, saving a manual cleanup step.
A good voice-to-text tool strips filler words and resolves an in-sentence correction without you having to edit the result afterward.

03Wispr Flow vs Built-In Dictation

Default OS dictation tends to transcribe literally, including every filler word and self-correction, leaving you to manually clean up before sending anything.
The gap between a literal transcript and a formatted, corrected one is the actual value voice tools compete on, not raw word accuracy alone.

04Prompting Gemini

Speaking a prompt out loud, rather than typing it, makes it easier to add specific detail (exact hours, exact address, a visual reference) because thinking out loud is faster than typing it all.
Correcting a misheard or misspelled term once, like a business name, can teach a voice tool that spelling going forward, so the fix compounds instead of repeating.
Reviewing dictated text before submitting it to an AI tool catches small errors that would otherwise carry into the AI's output.

05ChatGPT and Snippets

A voice tool that works identically across every app removes the cost of learning or switching setups per tool.
Saving a long, detailed prompt as a spoken-trigger snippet gets you maximum prompt detail with minimum repeated effort -- a short cue instead of retyping or re-describing it.
Reusing the same saved prompt via a snippet also makes AI output more consistent, since the input wording doesn't drift each time.

06Voice Prompting on Phone

On mobile, a bad built-in dictation experience doesn't just slow people down -- it makes them abandon tasks entirely, like never taking notes on the go.
A voice tool accurate enough to work at a whisper solves the social-awkwardness barrier to using dictation in public or shared spaces.
Formatted output, like an automatically numbered list, means voice input on a small screen can still produce structured, usable text without manual formatting.

Glossary

Terms worth knowing.

Wispr Flow: A voice-to-text app for desktop and mobile that transcribes speech and reformats it to match the context you're typing into, such as an email, chat prompt, or note.
Snippet: A saved block of text in Wispr Flow that gets inserted in full when the user speaks a short trigger phrase, instead of dictating the whole thing each time.

Resources

Things they pointed at.

00:22toolWispr Flow ↗

02:17toolGemini

04:06toolChatGPT

Quotables

Lines you could clip.

00:00

“If you're only using WhisperFlow as a super accurate voice to text tool, then you're only taking advantage of half of what makes it so incredible.”

clean thesis statement, no setup needed→ TikTok hook↗ Tweet quote

01:28

“Notice how it didn't just type out what I said. It knew I was writing an email, so it formatted it correctly for email.”

concrete demo payoff line→ IG reel cold open↗ Tweet quote

03:23

“Whisper Flow not only saves you time because it's super accurate, but also because it types what you mean, not just what you say.”

tight punchline summarizing the core differentiator→ newsletter pull-quote↗ Tweet quote

The Script

Word for word.

Read-along

Don't just watch it. Burn it in.

See every word as it's spoken — crank it to 2× and still catch all of it. The same dual-channel trick behind Amazon's Kindle + Audible.

If you're only using WhisperFlow as a super accurate voice to text tool, then you're only taking advantage of half of what makes it so incredible. Since I started using WhisperFlow, I found that I rarely ever type now.

Not just because WhisperFlow is four times faster than typing, but also because prompting is a 100 times better with WhisperFlow. It's kind of addicting.

So I wanna show you why that is and how you can use WhisperFlow to create better prompts and get better outputs from AI as a result. If you're not familiar with WhisperFlow yet, it's an AI powered voice to text tool for your computer and your phone. Before I show you how it works, I wanna let you know that this video is sponsored by WhisperFlow.

There will be a link in the description that will get you one month of WhisperFlow Pro for free. But I've been using WhisperFlow for months now on my own, so when they reach out to me about sponsoring this video, it felt like a natural fit.

For instance, when I write emails now, I don't ever type them. Instead, I activate Whisper Flow by holding down one button and start talking. Hey, Jamie.

Thanks for sending over the lesson. I took a look at it. I, um, think it mostly looks fine.

Um, however, I think we should swap lesson four with lesson five I mean, Let me know if that works.

All the best, Paul. Notice how it didn't just type out what I said. It knew I was writing an email, so it formatted it correctly for email.

It also removed all those ums and uhs and even fixed the correction I made mid sentence. If you try to do that with the built in dictation, we all know how different the result's going to be.

Hey, Jamie. Thanks for sending over the lesson. I took a look at it.

Um, I think it mostly looks fine. However, I think we should swap lesson six with lesson four.

No. I mean, lesson five. Let me know if that works all the best, Paul.

Look how messy that is and how it's not formatted for email. So Whisper Flow not only saves you time because it's super accurate, but also because it types what you mean, not just what you say.

But where WhisperFlow gets really addicting is when you use it to prompt your AI tools. Here I'm inside of Gemini and I wanna ask it to create a website for me, except I'm not totally certain yet what I want the website to look like. With Whisper Flow, I can just think out loud, let Whisper Flow type for me, and then Gemini will take care of everything and create the website I want.

Create a one page website for my new restaurant Luna Snack Shack. I, uh, wanted to have a cabin like feel to it with like pine trees on the page, maybe a bird. I mean, a loon somewhere.

Let's, uh, also include the hours of operation, which are from 11AM to 11PM every day.

Location is on Main Street in Longmont, Colorado, and include the menu. Let's just make it like traditional snack shack type food.

Before submitting this along to Gemini, I wanna point out a couple of things. First, of course, it cleaned everything up. But I also wanna point out that right here, the name of the restaurant is Luna Snack Shack, but this isn't spelled correctly.

I wanna do a little bit of a play on words here where I call it Luna with two o's as in a loon the bird. So I'm just going to go ahead and change this to Luna.

And what's happening in the background is that WhisperFlow notices I made that change and has now added that spelling to my dictionary.

So next time I say Luna Snack Shack, it's going to spell it correctly. And I didn't need to go into the settings and add that.

It all just automatically happens. So, of course, once I'm happy with this, I'll send it along. Gemini now takes over.

But as for the actual writing part of it, I never have to worry about it anymore. I just think out loud with WhisperFlow. Of course, WhisperFlow works anywhere you can type.

So if we switch over to ChatGPT, I can use WhisperFlow in here as well.

It works exactly the same and carries over all my settings and my dictionary. It also carries over my snippets. Snippets are shortcuts.

You say a short cue and WhisperFlow instantly drops in a full longer block of text. Here, for instance, I have a prompt I've saved as a snippet in WhisperFlow. Every time I say studio prompt snippet, it will drop in this prompt that I use to clean up images.

Let me show you how it works. So here inside of ChatGPT, I uploaded this image for myself, and I have this perfect prompt that I use to clean up my images and make them look nice.

Instead of typing that entire thing out, I can just hold down the Whisper Flow button and say, studio prompt snippet. And just like that, it fills in the entire prompt and I can send that along.

So notice how fast that was. Typing that out would have taken much longer. And as a result, I usually wouldn't give such detail in the prompts that I use.

But with WhisperFlow, it makes it very easy both with the snippets or even just talking out a prompt that I give it a lot more context and instructions much faster than I ever could with typing.

And with the snippets, the results are, of course, much more consistent because I'm using the same prompt every time. Where I find Whisper Flow really helpful is on my phone where it's difficult to type out long prompts. I just open the app I wanna use and using the Whisper Flow keyboard, I can whisper to it.

It's accurate enough to understand what I'm saying, but no one else around me can hear me. Can you give me some good workouts for bouldering when I'm at the gym and don't have access to a climbing gym?

I actually do a lot more on my phone now than I used to because of WhisperFlow. It has removed that roadblock where it just feels so difficult to type on here.

Now I could just talk using WhisperFlow. Even just doing simple things on my phone now like taking notes is so much better with WhisperFlow. In the past, because the built in dictation is so bad and would get things so wrong, I would never take notes on my phone.

If I had an idea, I would just try to remember it until I got home and I can get back to my laptop to type out the note. But now WhisperFlow perfectly writes and formats notes for me, and I could do it all with one hand.

Ideas for future videos. One, how to prompt better using WhisperFlow. Two, how to create any website using Gemini.

And three, the best way to prompt ChatGPT. Notice how it put it into this numbered list for me, perfectly formatted, everything is accurate, and I didn't even need to use my other hand all with one hand.

So this makes it so much easier to use my phone and and work on my business while I'm on the go. So that's how WhisperFlow got me addicted to voice prompting. It's faster, it's easier, and I get way better results with it.

If you wanna try WhisperFlow out for yourself, use the link in the description of this video to get one month of WhisperFlow Pro for free. It's free for a month, so you might as well try it out. I'll bet after just a few days, you'll get why voice prompting with WhisperFlow is so addicting.

Thanks so much for watching, and I'll see you in the next video. Bye for now.

The Hook

The bait, then the rug-pull.

A creator who's used Wispr Flow for months makes the case that voice-to-text isn't just a faster way to write -- it's a fundamentally better way to prompt AI, because talking naturally produces richer, more detailed instructions than typing ever does.

CTA Breakdown

How they asked for the click.

VERBAL ASK

00:22product

“Use the link in the description (or promo code LIPSKY at desktop checkout) for one free month of Wispr Flow Pro.”

Sponsor disclosure and CTA delivered plainly in the first 30 seconds, then repeated with an on-screen promo-code overlay in the outro -- low-pressure, no hard urgency language.

MENTIONED ON CAMERA

00:22toolWispr Flow ↗

FROM THE DESCRIPTION

PRIMARY CTAWhere the creator wants you to go next.

ref.wisprflow.ai ↗

Storyboard

Visual structure at a glance.

open

hookopen00:00

email demo

valueemail demo00:55

comparison

valuecomparison02:18

gemini prompt

valuegemini prompt02:18

CTA

ctaCTA07:04

Frame Gallery

Visual moments.

open

Frame at 00:09 from Wispr Flow Got Me Addicted to Voice Prompting

Frame at 00:16 from Wispr Flow Got Me Addicted to Voice Prompting

Frame at 00:21 from Wispr Flow Got Me Addicted to Voice Prompting

Frame at 00:25 from Wispr Flow Got Me Addicted to Voice Prompting

Frame at 00:31 from Wispr Flow Got Me Addicted to Voice Prompting

Watch next

More from this channel + related breakdowns.

14:08

Paul J Lipsky · Tutorial

How ChatGPT Sites Turns One Prompt Into a Hosted Website or Private Dashboard

A tour of ChatGPT's built-in site builder — from a class homework page to a private morning-briefing dashboard — built from nothing but plain-English prompts.

July 21st

19:07

Paul J Lipsky · Tutorial

Claude Just Changed Completely: Here's How It Works (In 2026)

A full settings-to-scheduled-tasks walkthrough of Claude's post-update interface — Chat, Cowork, Projects, Connectors, Skills, and the model tiers that decide what any of it costs.

July 15th

12:18

Paul J Lipsky · Tutorial

How to Prompt Claude Opus 5 (Anthropic's Official Rules)

Anthropic published an official prompting guide for Opus 5 — five rules, tested live by building a real website in under three minutes.

July 27th

11:14

Paul J Lipsky · Tutorial

NotebookLM's New Vertical Shorts Are INCREDIBLE

Paul J Lipsky tests NotebookLM's free new Shorts feature — 60-second vertical AI videos for everyone, powered by Google's cheapest image model yet.

July 1st

13:35

Paul J Lipsky · Tutorial

GPT 5.6 Edited This Entire Video

A YouTube tutorial on wiring ChatGPT and Claude into a $79 desktop editor called Borumi so they cut, zoom, and animate your talking-head videos for you.

August 19th

07:18

Paul J Lipsky · Review

Claude Opus 5: First Look, and How It Stacks Up Against Fable 5

A launch-day test of Anthropic's new default Opus model — benchmarks, a landing page build, a script draft, and a motion graphic, all run head-to-head against Fable 5.

July 24th