Show HN: Stable Diffusion print-on-demand apparel

IMHO, anyone who enjoys entering text via speech already uses it as a feature of their operating system (I know android and iOS both have dictation buttons). Most people don't use these features however, for good reason.

I wanted to try it out, but unfortunately I don't even have a microphone on my current device.

I just want to see some example prompt/output combinations!

Add the ability to write a text prompt; accessibility wise - not everyone speaks the same or can be understood by basic speech to text

Yup. After 6 attempts it still couldn't understand me. And I'm a native english speaker, grew up in the US northeast, and nobody would accuse me of having any sort of non-US-broadcast-standard accent.

And some of us just plain don't want to use voice interfaces.

I don’t mind using them, but this one doesn’t work. It makes for some comical images but this is not an effective voice prompt.

The idea of using stable diffusion to create a t-shirt is compelling, but the execution here leaves a lot to be desired.

Randomly, I picked the stencil style and went for "Man with beard base jumping from the Eiffel tower." It produced one really poor image. It wasn't stenciled looking at all.

Next I went for pop art and used the prompt "Frog riding a motorcycle". The art style looked like something that an 8 year old would draw, not Andy Warhol (which I believe is the artist in your screenshot).

So I think you have some work to do in taking the inputs from the user and crafting them into better SD prompts that generate outputs that look like t-shirt artwork and in the styles they're supposed to be. I'd also drop the voice prompt support, no added value here. Definitely would prefer to type it on my computer and mobile keyboards already have a voice-input button.

It's easier to get good results by generating 2-8 images for each prompt, picking the best result, using that as the starting point for a new run.

Tweak the parameters of how close the new image should be to the old one and the prompts, tweak the prompts to address problem areas in the image, generate another 2-8. Rinse and repeat until you have something decent.

I doubt that the really impressive SD images are coming from submitting a prompt once and taking that output. It's better than past open-source efforts like the mini dall-es, but it also still has a way to go before it can reliably produce good results on its own.

AI art on products is an interesting idea, but personally I would use a more fully-featued SD web UI to generate the image locally.

This is one of those things where op needs to do something similar to what I think midjourney is doing and do prompt optimization.

They need to do some research into what additional prompt cues they can add for the user automatically to bring the user's prompt quality up.

They can append these to the prompt behind the scenes or provide a second text input of the additional prompt data to allow the user to modify that separately from their initial prompt input.

It might even help the user better understand how styling works with prompts.

Many complained about voice-only input. I want to highlight a few other aspects that could be improved:

1. Navigation doesn’t work. Sometimes swiping right goes to the next page, sometimes doesn’t. Then you are stuck on the 3rd page and can’t go back.

2. What are points? I have 5 points, got to the payment step - the T-shirt is $25. How am I supposed to use points?

3. The T-shirt that is displayed as a 3D background is completely different from the low quality generic fit design at the end. This discrepancy is off-putting.

4. The tiny final print is unintelligible on my phone. Some squiggles and circles that I can’t make sense of at this resolution/zoom level.

I think the idea of custom fashion is interesting, but this implementation doesn’t feel empowering at all. Feels limiting and gimmicky. More importantly, the result - a generic T-shirt with a very small print is not fashionable or even attractive.

This reminds me of my failed business idea https://indivicia.com/

The problem with printed shirts is that everyone expects them to be cheap-ish. And the Printful quality is low enough that people were upset at a $50 price point. The only tshirts that sell for higher prices are branded ones, e.g. with Marvel characters on them. But then you're effectively selling copyright infringement as a service. And your profit margins will be low overall because Printful and their shipping are rather expensive.

In the end, and with external help, I then acquired and ran https://meowbikinis.com/ for a while. Profit margins were insanely better. Until the lockdowns in China started and the last leftovers of my stock spoiled inside a rented fulfillment center...

have to say - meowbikinis is super interesting story. Can you share a bit more about it?

The input does not do speech to text well. I would like to use this service. Let me type my prompt.

I tried 5 times and every time it messed up what I said. My os has a good speech to text tool and I would use that if I wanted to talk a prompt.

The site is cool and it’s a neat idea, but forcing microphone use is weird and ineffective.

I expected it to somehow fill the entire shirt, but apparently it just makes a rectangular patch of the image on a plain black/white shirt. Also, the speech to text doesn't seem to do that well.

The rectangle is probably a limitation of the t shirt printer service they're using, I've also seen that they don't usually fill the entire shirt.

They could print on the entire shirt as that is an option when they setup the store but the profit margin would make it too expensive for a third-party.

That would require the more involved 'cut and sew' process.

Sorry, someone already did it https://www.thisshirtexists.com/ Related HN launch https://news.ycombinator.com/item?id=33082472

Why do you need microphone access? Can't you use text instead? I'm not giving mic access to some random website.

they just said because you can speak the prompt. kids these days dont want to use their hands, that's for babies.

https://youtu.be/KMy1zO8m8sM

I saw that. Not exactly a great reason.

What's the reason for the "everywhere except Europe" shipping? I'm guessing some legal/regulatory thing, but would be curious to know the details.

As stated in the original post printing and shipping is handled by https://www.printful.com/ and they offer their services in Europe.

I assume it has something to do with copyright regulations in Europe or similar?

GDPR maybe if the microphone data is saved?

Great idea - I actually did this manually already by generating an image and uploading to vistaprint. One thing I noticed was that it’s hard to generate images with no background so you could consider automatic background removal as a value add over the standard image generation algorithm.

Thanks for the mentions everyone. I've already launched text2shirt some weeks ago - no need for audio, online and serving shirts and hoodies:

www.thisshirtexists.com

Hey Matt ... would like to connect if you are still monitoring this thread.

Not giving microphone access unless it’s something _I_ need, not the website.

Would be great if you could add text input as an alternative to voice input.

>Please note at this early stage we only offer shipping to countries outside of Europe.

Don't you mean inside?

Probably due to EU tax law. I did the same when I was using Printful with a German company. Shipping to the US is cheap and tax free, shipping to the EU is annoying.

Cool idea, but my first and only attempt returned "something went wrong".

> You have denied audio permission. If you want, you can always go to your 'settings' and allow it to fully experience this site

Weird phrasing. If I don't want to I can't experience the site at all.

An app being confusingly passive is the blinking 12:00 display of our time.

For those that want a text box, I've used https://thisshirtexists.com, works well.

Insane that OP's site works with audio and only audio. I mean, who thought the more complicated way to enter text should be not only the default, but the only option at all?

Presumably they’re working on an alternative using device accelerometers to allow Morse code input.

Having a voice recognition interface for something I could type (more accurately) in 10 seconds is very off-putting. I don't use Siri for a reason.

Really, really bad idea to use force using microphone. Unreal.

Perhaps it’s just a honeypot to capture some voices of hacker news readers.

I’m now suspicious whenever someone has a stupid tool that the real purpose is to train some model with the input.

Instead of "you're doing it wrong", suggest alternatives. When someone is learning, help them learn more.

When something isn't good, you needn't pretend that it is, but don't be gratuitously negative.

https://news.ycombinator.com/showhn.html

The alternative here is pretty obvious - a text box.

'Don't be a jerk' is the more important, seemingly less obvious alternative.

Requesting microphone permissions for something that would actually be easier without it is also very obnoxious, and for anywhere with GDPR perhaps even illegal if the data was persisted. Maybe this is why it isn't available in Europe?

The guideline is there to avoid escalating obnoxiousness because it trashes the forum.

As opposed to pulling this sh!t? https://news.ycombinator.com/item?id=33494109

You're going to get sued in the US if you try to market this thing and require voice input. This is potentially a huge accessibility issue.

You've essentially eliminated (most) people who are deaf and people with speech issues from being able to use your product.

It's a neat idea and the design looks like it could be nice but I can't really get past the first page because I cannot use it (no mic attached to this device and I wouldn't want to grant you permission to it anyway.)

I have no microphone attached to my computer, so I can't use your AI to order a t-shirt. What a weird sentence.

An important question to ask yourself before you launch a product: when you look at the final result, do you personally think that it delivers on your value prop? If you don't think so, then users, who are often more critical, definitely won't either.

"you don't have to be a fashion designer or graphic artist to create amazing clothes that express your personality. Whether you want to print a simple design or something more complex, 1SEWN makes it easy and affordable to get the perfect item, every time"

Do you think a generic white T-shirt with a small print of a low artistic quality is really expressing anyone's personality? Or that it looks like a "perfect item"?

We are living in the future. In the next iteration, I expect direct connection from my Neuralink to the Creative gods of AI Metaverse. Why speak at all. This is for babies.:)

> map the outputs onto actual physical products

But wouldn't that be more like an actual 3D printed cat instead of a picture of a cat on a t-shirt?

> Shipping is currently limited to outside Europe

So you ship to North Korea but not Germany? Got it.

Taking fast fashion to a whole new level. Depressing.

This isn’t inherently fast fashion, and on-demand shirt printing isn’t a new thing. I had a cafepress t-shirt I used for ~10 years and only got rid of it because I lost weight.