[sjv]

Non-statistical methods of imitating handwriting are always fairly easy to identify as synthetic due to the uniformity of strokes and unnatural kerning (among other things). Even methods which incorporate random variations don't hide these artifacts sufficiently well. In my opinion, the only way to generate synthetic handwriting which is convincingly real is to use statistical methods (i.e. machine learning) to model all variation which is present in real handwriting. I've implemented a neural network in javascript which does exactly this - https://seanvasquez.com/handwriting-generation/. You can play around with it and find a few weaknesses, but in general I find that it can produce handwriting which is indistinguishable from real handwriting.

[crazygringo]

That's super-cool. Honestly that's what I was expecting the submission to be.

When I looked at the example image in the original submission, the same letters all look basically identical to me -- I can't figure out why it's JavaScript rather than just a font.

Yours actually looks like real handwriting. It's quite slow, though -- like the actual speed it takes to write. Was just wondering if you slowed it down intentionally for effect, or if that's just how long the NN takes?

Also, I'm not familiar with the space, but this feels like something you could commercialize as a plugin for Illustrator or something like that -- particularly if you had a range of sliders to really fine-tune the desired text properties (width, weight, weight variation, etc.).

[sjv]

It's not intentionally slowed down. The sampling process is computationally heavy and you'll notice improved performance when running it on better hardware. There are many algorithmic ways to speed up neural network computation and I'm fairly sure I could speed things up by an order of magnitude or two if I invested some time into it.

[pkphilip]

You sure you don't have someone from India writing this out on a touchscreen when someone enters the text at the prompt?

(Just kidding)

[MikeTheGreat]

This is really neat! I love how it looks, and the selection of styles.

A couple of times, using Style #2, it generated a bunch of loops at the end:

Text: "The quick brown fox, etc" Image: https://pasteboard.co/Jn0O17a.png

Still - nifty, and fun to play with. Makes me want to find an application / excuse to use it :)

[aldanor]

Here's a "Hello World" with a whole ton of loops as well :)

https://imgur.com/fhG2Fwq

[chrisco255]

That's a totally acceptable "helloooooooo world"!

[sjv]

The model was trained on a relatively small dataset which has few occurrences of the letter x, almost none of which are in cursive. As you've found, asking the model to write out an x in a cursive style can totally derail the generation process.

[fps_doug]

Any ELI5 for people who know close to nothing about NN, AI, machine learning etc?

Like, how does one letter it doesn't know how to write well make it skip/mess up the following ones too? I mean, "derail" seems a fitting description to what I'm seeing, but how?

[alexanderskates]

The "handwriting" data for this model is basically the coordinates of a pen. The length of the string representation of the text is very different from the length of the coordinate representation of the text, therefore the model "learns" a window corresponding to when it is drawing the current letter, and when to start the next letter. For these letters, as the model doesn't learn how long this window should be, nor how to transition from it to the next letter, it gets stuck and outputs nonsense.

[chris_st]

There must not have been many "z"s either -- "quartz" failed as much as "fox" in my tries. Still, nice work!

[iforgotpassword]

I got that to after playing around a bit and then trying that phrase. It gave me a really good laugh, as if AI just went "ah shit, why does everyone enter this stupid phrase eventually?"

[vibragiel]

It breaks down catastrophically with diacritics https://imgur.com/8U3IKR9

[hombre_fatal]

Heh, "¿Cómo estás?" -> "OC"

[maneesh]

Wow! How did you do this? Any tutorials or methods that would be helpful to learn how to produce something like that?

[gfxgirl]

From the github project

https://arxiv.org/abs/1308.0850

https://github.com/sjvasquez/handwriting-synthesis

[mushishi]

Hi, I like this, and found your project at Github. Could you add licence to your project, please! https://github.com/sjvasquez/handwriting-synthesis

[ldd]

I second this. Please add a license!

[detritus]

Wonderful! And as it generates single stroke SVG exports, I can output to a laser cutter and have it write for me :)

I can't think of a practical use for this, but that's not stopped me wasting the last 15 minutes getting my machine to write out expletives in neater cursive than I could manage...

- ed

Oh, and this is by far the best digital 'handwriting' I've ever seen, despite the occasional glitches and believe me - I've been interested in this sort of thing for decades.

[detritus]

Having played with this a bit - if you can flesh it out a bit, I am sure you can monetize this.

I've been dicking around with it for an hour or two and I'm getting results that - if i mingle outputs - are to my eye completely human in nature.

Honestly, as an ex sort of still designer who still engages with that sort of community, i see a tool like this that makes neat scripty vector paths (and I think that is particularly important) as being super-useful. It outputs unique forms that are individual in nature. I'm thinking all the graphic designers who makes stuff for weddings and invites and whatnot.

Needs work, and I have absolutely no concept of how complex it is to refine things, but what you have here is a helluva start to something I think could be immensely useful to some people.

- minor eds

- ed ed

Can you manage the stroke widths through the cursive forms, thinner when drawn quicker and straighter, fatter when slower around corners, for example?

[axaxs]

Does this not work on mobile? I just get random lines then they disappear.

[kristopolous]

Working on Android chrome for me, a little cludgy but tolerable

[axaxs]

just checked again and sure enough it's working now, thanks.

[richrichardsson]

Very interesting. I did notice that with my input of "Hello, this is some hand written text!", the output of the word "text" consistently looked more like "tent" than text, regardless the style or legibility settings.

[laszlokorte]

No matter what style/legibility/text I try the line always just drops as if my computer had a faint [1]

[1]: https://imgur.com/a/IgAV1ef

[Kosirich]

If I write for example 5 letter "k" in a row (space separator), the technique used to write them varies in a way that irl wouldn't for a real human.I guess you could argue why would someone write 5 k letters. (I won't mention that I observed the same variation when repeating the process using a name of a certain notorious organization that uses 3 k letters)

[detritus]

How would you go about creating a minimal/medium source of input to train this?

Could you create a 'written test' type thing - a couple of pages of selected forms of repeated instances of glyphs and interactons - that would be used as input and to learn a passable subset of one's own handwriting?

[sbarre]

This is super cool!

A random FYI: the demo seems to consistently fail on the letter x when using style #9 and typing out:

The quick brown fox jumps over the lazy dog

I tried a few different spots on the legibility slider with similar results..

It draws the first "top left to bottom right" line, but then instead of a "top right to bottom left" it draws another "top left to bottom right" next to it..

[TheNewAndy]

Try "fox" in style 2 with maximum legibility. It goes loop-the-loop several times.

[glerk]

Seeing this had a strong uncanny valley like effect on me. Like my brain can't help but imagine a sentient agent writing the text and suddenly becoming psychotic.

[TheNewAndy]

hmmm, not necessarily the same output each time

[pdamoc]

This looks very interesting. I wonder how much data would be required in order to imitate a particular handwriting style? i.e. if I know someone with a lovely penmanship and I can ask that person to create such a reference data set, how much should that person write in order for the network to be able to reproduce the style reliably?

[habosa]

Yours is amazing! By the way it has a big problem with "/" in the default settings. I tried doing a date like "08/20/2020" since that's something I need for signing forms but the slashes mangled it. Dashes worked ok.

[playpause]

This is amazing. It's cool how you can drag the legibility slider up and down while it's writing and it responds live. It looks the most realistic to me when I alternate between high and low legibility, especially with certain styles like #3.

[Waterfall]

Is there one that instead imitates pressure sensitive input and then follows the same pattern? With a sufficiently large brush and downsizing the image it could look very good. Maybe the lack of artifacts from the brush is the reason for the unnatural look

[aketchum]

How much data is needed to train the model to replicate a specific person's handwriting? Could this recreate my hand writing with less than a page of notes in my own handwriting?

[navanchauhan]

It’s giving blank output on iOS or is it because the model takes time to load?

Also, is there any chance we can fine tune the model for each user by taking their handwriting in a predetermined template file?

[jsjohnst]

Works for me on iOS. Did you click the “write” button in the bottom right corner? It seems you have to tap it after any change in text or settings for it to update.

[navanchauhan]

Yup, I did use the write button. It works on my laptop though.

I checked the console and it was indeed throwing an error

"Error: Backend name 'webgl' not found in registry"

[jsjohnst]

Were you using actual Safari or an app’s stripped down UIWebView? If Safari, then I’m guessing you have a content blocker that’s stopping access to WebGL.

[navanchauhan]

The actual browser.

I am only running Pi-Hole on my network but then again if that was a problem it would not have worked on my MacBook either

[fizixer]

Thanks for sharing. If you don't mind, did you use Graves 2013 RNN or something recent/more advanced?

[sjv]

Yes, it's mostly based on that. Primary difference is that this model adds an inference model (VAE style), which allows you to a) sample style vectors from a latent distribution b) do so more efficiently than the priming mechanism described in that paper.

[fizixer]

Thanks. Great work.

[adventured]

That's a lot of fun to play with. Nicely done.