Episode Transcript
[00:00:05] Welcome to Prompting Curiosity, a podcast for the AI curious. No coding background required. I'm your host, Dr. Shantae Cofield, also known as the Maestro, and I created this show to explore what these AI tools actually are. Really, though, are the files in the computer, how to use them, and what they might mean for how we think, work, create, and move through life. Whether you're skeptical, intrigued, or already experimenting, you're in the right place. All that I ask is that you stay curious. All right, let's get into it.
[00:00:38] Hello, hello, hello, my curious people, and welcome to episode 45 of Prompting Curiosity. I am your grateful host, the Maestro, and today we are talking about one of my favorite AI powered tools, Whisper Flow. So, uh, I know, as I say, AI, uh, powered. Maybe you're rolling your eyes. I'm guessing not because you're listening to this podcast. But in general, everything got AI in it. We don't want it in anything but this. I actually really do like having it. So one of the things that I love about having this podcast and why I'm happy I changed the name is because I can talk about all things AI. I was doing that before anyway, but now it kind of fits a little bit better. And I don't have to, like, specifically talk about, you know, a specific type of AI or like a single company.
[00:01:25] Um, my goal is always just to share what I'm doing, what I'm using, what I'm thinking, and then you curious folks are free to take what serves you and leave the rest. So what I have been using for quite some time now, and I've mentioned it in past episodes, I think. Um, I don't know. I talk a lot. I know I definitely mentioned it on my vlog that I am, um, just insanely behind on.
[00:01:48] Uh, but what I've been using is a thing product tool called Whisper Flow. It's spelled W, I, S, P, R, space flow, F, L, O, W.
[00:02:00] Um, so first off, I do have an affiliate link. No, they're not paying me to say this or to do this podcast. I just really like the product. But I have thought about pitching them, uh, but I haven't pitched them. And you'll hear my. My hesitation later on that episode. But the affiliate link will be linked in the show notes. Uh, if you use it, you can you get a free month of pro if you sign up. And, uh, then when you hit your first 2,000 words, I get a free month. So zero pressure to use it. Zero pressure. Any. There's never any pressure. But Mama didn't raise no fool. So I'mma drop that affiliate link. Feel free to use it would love that. Um, but let's get back to the actual product. Whisper Flow. What is it? It is an AI powered dictation software. Perhaps you're saying, maestro, that already exists. I got talk to text already on my phone. Yes, it does. I. But have you tried it? Because it sucks. At least for me. You all know I be speaking hella fast and Siri slow, she can't keep up. And then she'd be just writing crazy things and I'm like, what, what's going on? So my friend K, he put me onto Whisper Flow last December when I was getting into all the AI, ah, coding stuff. He had been using it for a bit. He does a lot of Descript, uh, videos and he was sharing them with me and then he would be talking to the computer and I'm like, oh, what are you using? That's like, great. You don't have to, you don't have to type. And sometimes that's really helpful when you don't have the words and you're just kind of like riffing, right? And so it's nice to like just speak. Uh, and he was like, I use this software called Whisper Flow. And I was like, let me try that. And I have not looked back. Right. I, I love it. Um, like I said, it is AI powered, right? It is AI powered voice dictation. You press a button or like a hotkey hotkey, meaning you like program a key to do what you so turn, start recording basically, right? You press a button, you just speak, you just talk whatever you want to say and it inserts the clean polished text.
[00:03:56] Right? Ever wherever. Right Ever wherever your cursor is, whether you're in Gmail, you're in Notion. I don't fucking use Notion. But that's one of the things they talk about on, on the, the marketing for it. If you're in Slack. I don't use Slack either. If you're on a website, if you are on your, on your phone wherever, it doesn't matter.
[00:04:14] Anywhere that there's a cursor you can use, uh, Whisper Flow. I stumbled for a second there because there's another thing called cursor that's completely different. Um, but anywhere that, where there is a cursor you can use Whisper Flow of note. You don't have to hold the button down either. On your phone. You just press the button and it starts, and it starts listening and then you hit stop and then it pastes it in that to me is kind of like a little bit annoying. Um, um, because you won't see it like being written in real time, but it doesn't lose anything. You're saying it like saves it locally and, and then it can save it to a, um, like back it up to their, to their servers. You get to choose with all of this, um, because there's privacy settings. But all that to say you don't have to hold the button down. You press a button, you start talking. When you're done, you press the button again. On your computer you have the option of holding the button down, which is annoying. Or you can set it up so that you have a hotkey shortcut and then you press those. So I have it as like the little globe and the space bar. When I push those together it starts recording and then when I push the little uh, like globe then it stops recording.
[00:05:23] Or I could just hold the globe down the whole time as I'm talking. It just depends on what I, what I'm doing. But hands free is typically the way to go when you're dictating. So a little background about the company and why I don't. I haven't pitched them to sponsor the podcast. Besides the fact that I have had my other podcast for you know, 11 billion years when, since what 2018.
[00:05:44] And uh, I've got to go with commercials on that. I kind of want, I don't want to do that for this one but wouldn't help to you know, get some money. But uh, Whisper Flow was founded in 2021 by these names. I don't want to butcher them but today Kathari and Sahaj Garj, um, two individuals out of San Francisco. They originally built this software to be a uh, wearable that let you control your phone without touch.
[00:06:13] They pivoted to software into this software and they launched Flow in 2024. Uh, people be iterating and having money and, and doing all sorts of. But where does the AI come in? So your speech, when you talk it runs through cloud based AI models that understand the context of what you're saying.
[00:06:32] Uh, uh, these models are unfortunately open A models and you, you folks know I moved away from, from OpenAI.
[00:06:39] Uh, and so I'm kind of like oh, uh, yes, I'm still using Whisper Flow. I, I am not perfect. Like give me, you know, cut me some slack there. Um, but I haven't pitched them at least to sponsor the podcast. But the model that it uses is an Open a model aka ChatGPT.
[00:06:54] Um, um, when you speak. The AI automatically removes any of the filler words. It fixes sentence structure, it adds punctuation. It formats the output app that you are in.
[00:07:05] So it cleans up the text as it's going and as you're speaking as opposed to performing just raw transcription. Exactly what you say gets documented. Um, which is really nice. You, there's different settings for how much you wanted to clean up. I just wanted to, to write down what I'm saying. But if you have a tendency to say a lot of ums, it'll take care of that.
[00:07:26] Uh, I like that it will add the punctuation and everything and just cleanly. Um, it works in 100 plus languages. And yes, as the name implies, it works when you're whispering. I can have the mic far away from me. I have the phone close or near. And it works like, and it picks it up, it gets it right. And that's, that's what I want. I want it to be correct. So that's a good segue into the next part, which is what makes it better than the native dictation is that it gets it right. Right. If you've ever used talk to text in Apple Notes, you know, the struggle is real. It should be just messing up and like it's bad.
[00:08:05] Especially uh, if you know, you're using it to take a quick note and you're like, oh, go back to it later. And you go to later, you're like, what the. Did I say? What is it? What is even going on here? Right? And it definitely doesn't, you know, end the sentence as well or format anything for you.
[00:08:17] And one of the things that Whisper Flow likes to promote and tout is that it cleans up.
[00:08:23] You know what, it cleans up your, your, your speech as you're, as you're dictating, right? So Apple Google, the native dictation software that's on your phone and such. It's literal. It types exactly what you say, filler words and all. Whisper Flow removes those arms. It removes those, uh, things like that.
[00:08:42] Um, the, again, the most awesome feature to me is handling the punctuation happening, handling the capitalization. And it will automatically format things into a bulleted list. If you're reading item list format. M. That's really cool to me. I haven't had a case yet where I'm like, I wish that was just in a regular sentence. It, it, it, it's. If I'm saying it in a list as I want it in a list, it's been super. This is super. Cool. To me, I think it's a really useful feature. Um, it will learn your words and phrases and your tone.
[00:09:11] It'll keep it consistent across any of the apps because you can use on your phone, you can use it on your computer, wherever. You can use it on your, on your laptop, wherever you install it. Um, and then there is a personal dictionary feature where you can add if you have jargon, brand name, whatever, uncommon words, your own names. Shantae can be difficult. Um, you know, it's not like it's naturally in a lot of these dictionaries. So.
[00:09:31] Because racist. But, uh, yeah, this is this cool features there. But flip side, as with all things, there are features that in my opinion are not super necessary. And like, these companies feel like they have to keep iterating and keep like adding. And I'm like, just make sure that it works well. Like that's all I care about. But, uh, they do have what's called command mode, which where you can highlight text and say, make this more formal or turn this into a bulleted, you know, bulleted list or summarize this and it will do that. Which, um, is cool. I guess it does have style modes as the ones that I've seen as of right now is the different styles are formal, casual, very casual and excited. Um, and this means that you, when you have that mode selected, it shapes the out the output tone. And so for me, my voice is my voice and I speak the same on all platforms. I don't need to change, you know, how I'm writing this email versus how I'm, you know, making a note to myself versus how I'm making a caption for Instagram. It's all the same. Um, but if you do personally need to be switching, then it's an option. So my honest opinion, I definitely use it a lot more when I'm at my desk. Not so much on mobile for general texting because you have to do that annoying switching of screens, I don't think. I actually don't know if I said that earlier.
[00:10:41] Um, with the iOS update, you have to open it like there is a. You add it as a keyboard, there is a button, you push it, but then it opens in a different screen and then you have to toggle back, uh, you know, move back to switch, switch back to the actual texting screen, which I find annoying. Um, so I don't really use it a ton on mobile. Also, like, I don't need to be making that long of a text. Um, but I will use it on mobile if I'm dictating things that I'm writing, like notes or blog posts or things like that one that I do use my phone for, and if I'm out on a walk or something. So there is a use case there. But, um, typ. Typically I'm using it at my desk. It is definitely faster than typing. But more than that, in my opinion, more than it being faster, is that it's helpful or maybe because it's faster. It is helpful for getting ideas out because we, uh, know. I mean, at least it's been my experience that oftentimes it's just easier to just speak, right. And articulate them by speaking and rambling as opposed to trying to type. Because our brains are like, I have to like, correct this as I'm going and make it make sense, whereas when you're talking, you can just let it rip. So that is, to me, the, the biggest utility there. Whisper Flow, the company, their marketing, they claim that it's four times faster than typing, which I don't know about, but it's definitely faster. And they base it, I'm just saying, numbers. I like transparency and honesty. Right. And things being realistic. So they base this Forex claim off of, you know, as per. It's on their like, homepage.
[00:12:08] They bases it off of typing 45 words per minute, which is kind of slow. Sorry if that's you. I'm not like shaming you, but, uh, that's kind of slow. And then speaking 220 words per minute, which is actually really fast. So I took a test. I will link that in the show notes as well. If you want to take a typing test I took, I know that I type pretty fast. And the results where I said it, I did like three times. And it was like, I type about 80 words per minute. 8, 0.
[00:12:31] And then there is a little dashboard because they be adding shit all the time. Um, there's a little dashboard that you can go to that like shows you your Whisper Flow insights.
[00:12:40] And according to that, I speak at 174 words per minute. And I am in the top 0.1%.
[00:12:50] So this claim that people speak at, uh, 220 words per minute, I'm like, I speak fast. I. I know I speak very fast, especially when I'm not trying to speak to another human. I'm just speaking to the machine. I'm like, it can keep up. Uh, and I'm only at 174, so I don't know about that. Um, but, you know, typing 80 words per minute and then speaking at 174, you do that math, it's like twice, you know, twice as fast.
[00:13:18] Um, but still it's definitely faster. Right.
[00:13:23] So last thing, it is in my opinion worth pointing out that typing and talking are different forms of communication. Right. And I'm not here to like replace one with the other and be like just be more efficient and more productive and faster for the sake of being faster. No, I said earlier I like to use the dictation software just to get things out of my head and then I'm going to be typing. Um, when I do these, the, the outline for this, I do not dictate it. I type it because it's annoying to dictate it and then actually say the episode again. Um, I type it and it is just, I don't know, it's the different part of your brain there. Um, so, you know, my choice with these communication means is to do both. Right. I'm going to write some, I'm going to type some.
[00:14:04] I'm not really writing much. I have like a whiteboard over here. But I'm not writing that much. But I'm going to type some stuff. I'm going to do talk to, text, um, for other things. But I just wanted to use this episode to, to show you a dictation software alternative that I have found to be really helpful.
[00:14:19] Um, that's it. All right, last section and then we will wrap it up. Um, how I use AI this week. If you're new here, Egypt, welcome. If you're new, welcome. Uh, each episode I share a quick example of how I used AI that week. So this week I used Gemini. I went over to Gemini, I did, uh, to create a video for the hero section of a website mock up that I did for one of my messaging clients. So, uh, you can just type in whatever product, whatever video you wanted to make and it will make it. It takes only a few minutes. Very quick. Um, I made three iterations of it because it made the first one and I was like, okay, everybody's naked, basically they have clothes on. Don't get me wrong. But this was for like a fitness video and like, you know, the clothes were so small and I was like, make the people have full clothes on and like look normal, not like an 18 million pack. So, um, I ended up doing three iterations. Each one I think was like 10 seconds long.
[00:15:11] Um, and then it said that I needed to upgrade and pay more money or I wait 24 hours for the usage to reset. I am on the pro plan. I don't know what the video generation, um, limit, I guess, is for the free plan. But I am on a pro plan for Gemini because I pay for Google Workspace, and so it's part of that. Um, but the videos were clearly AI, but they were also really good. Like, it works for the. Before what I needed it for, for just this mockup to show the, the client what I was thinking.
[00:15:41] Um, it. And just for such a simple prompt, right? It's not like I had to do all this crazy stuff. Like, very, very simple prompt. And it made the video, like, super, super simple. I'd have to go into, like a special part of Gemini, either just open and open Gemini and wrote it in the prompt and in the little prompt field.
[00:16:00] And it made it so very helpful use case, in my opinion. But I also, I'm not gonna lie, I am definitely very, very, very, extremely concerned about AI, uh, video generation. Right? And I, you know, I'd honestly prefer it not be a thing at all.
[00:16:20] Uh, but here we are. And to me, it's kind of. I mean, we've already seen it by swaying elections and just old people fall for it and like, how you can't tell anything anymore. You, like, literally don't know anymore. Like, is this AI or is it not? The thing that I made, it was very clearly AI, but there's so much on the Internet now and from all these other models, and you're like, I don't know actually if that is AI or if it's not. And to me, that is very, very, very problematic. So, you know, my push will, as with all things AI, will, will continue to be for more rules and more regulations. Um, but yes, that is how I used AI this week.
[00:16:58] Um, that is all for today, my friends. Hopefully you found this episode helpful.
[00:17:04] Uh, and if you did consider sharing it with somebody, you know, who's curious about the old AIs and perhaps could use some, maybe some new dictation software. Don't, uh, forget I got that Whisper Flow affiliate link in the show notes. If, if you do decide to try Whisper Flow, um, that's a free month for you and a free month for me once you hit those 2,000 words. So I'm trying to think. I'm going to look at my little, My little insight, uh, dashboard. I have at this point dictated a total of 241, 143 words.
[00:17:39] That apparently is two complete books.
[00:17:45] That's a lot.
[00:17:46] That's a lot. Uh, so, yeah, it's helpful for me. Maybe it'll be up for you if not, I'm just grateful for you listening.
[00:17:53] Don't forget. Don't forget.
[00:17:55] I, uh, have a companion newsletter, the Curious Companion, that drops every Thursday. That is basically the podcast episode. And by basically, I mean exactly the podcast episode in text format. So if you prefer to read or you just want a written record of things, you can join the newsletter, fam you head to prompting curiosity.com forward/newsletter or check out the link in the show notes.
[00:18:19] As always, endlessly, endlessly. I'm gonna say it one more time.
[00:18:25] Endlessly appreciative for every single one of you. Until we chat again next Thursday, stay curious.
