HN user
Founder of Handwriting OCR (https://www.handwritingocr.com)
This wrote a complete Laravel feature for me from start to finish almost perfectly, better than any other approach I have tried (e.g. TaskMaster). Amazing job.
I need to take the time to see what the Laravel side of this is so perhaps I can adapt for my RN app too.
Using this now - works brilliantly! What makes it Laravel specific? Would also love to use this with a React Native app I'm struggling to build.
From https://www.handwritingocr.com - seemed to be more accurate, mostly getting the New Mexican and possibly other parts:
---
and I don't know whether it can be reset for a date in December or not. Cornell seemed anxious that it should not come off too close to Christmas, and of course New Mexican practice would support him. I will take this up with the Judge and with Hanna the moment I can get rid of the brief. Meanwhile I would like to know whether there is anything else in which I can be useful to you, since it behooves me to be diligent in view of uncomfortable relations with the present management.
Are you going East in December?
Has any word come from Hagerman?
Were there any noteworthy developments at the hearings on the Tenorio tract?
I have no intimation yet whether I will be wanted in Washington, but the chances are that I will not.
With regards to all the brethren and flock, Dan
Very sincerely yours, George H. H. Baser
I'm the founder of handwritingocr.com - have you checked out our free trial? We have loads of educators using our service for exactly this, and they seem quite happy with it.
I understand your concern, and it's a common one. However, we can only give assurances in our privacy policy that your data is used only to perform the OCR, and nothing else. You can delete all data from the server immediately after downloading your results and no trace will be left.
Of course a local solution like Ollama is preferable for privacy reasons but, for now, the OCR performance of available local models is just not very good, especially from handwritten documents. With a couple thousand pages of journal entries, that means a lot of post-processing and editing.
I'd love to hear how Handwriting OCR (https://www.handwritingocr.com) compares for your task.
It's not free, but its accuracy for for handwritten documents is the best out there (I am the founder, so am biased, but I'm really excited about where the accuracy is now). It could save you time and for your 100 page project would cost only $12.
Hi,
There's no way to fine tune at present, but it does pretty well with all but the scrawliest of handwriting out of the box.
btw I wrote to the email in your profile, not sure if you got it.
Thanks, I'm grateful for this useful feedback.
Hey,
Thanks for the feedback and great suggestions. It's something I will try to implement in the coming days.
p.s. I tried to make my affiliation clear - I wrote "I’m the founder" in the original comment above :)
True, in cases like OP a one-off purchase will be better. But I also have business customers with regular and ongoing document processing needs for whom a subscription does work.
I expect the answer may be a combination of the two.
Thanks!
I’m still experimenting with pricing, and agree that per page pricing makes logical sense. Still, it’s harder for me to build a sustainable business on that model.
I will probably test a few per-page or single payment options soon, though.
Have you tried https://www.handwritingOCR.com?
It is designed to do exactly what you are looking for, and has been used very successfully by many others for that same purpose (I’m the founder).
It is not as cheap per page as Google Document AI, for example, but it does tend to be much more accurate for handwriting, so usually ends up cheaper when editing time is factored in.
If you find it does work well with your handwriting, please get in touch and I can try to fit the pricing to your use case.
Thanks!
Sounds like a challenge! https://www.handwritingocr.com :)
Hah, yes. It's just the remnants of a template. As I said - totally half-baked and not ready for (or expecting) users yet.
I left my last job and a toxic boss a few months back and decided the only viable future is to start my own thing again. But this time I was going to do it properly - lean customer development, speak to users before writing a line of code, etc.
Instead, I became fascinated by the massive improvements in OCR and how they can be applied to digitising boring paper records. Decided to start out with the most boring problem of them all - credit card and bank statements.
I decided this would also be a great project to make the leap to Python and Django. So, without speaking to a single potential customer, I threw myself into coding. After a while, I realised Laravel is so much better and jumped back to that. Have learned loads as a result.
I've built a local prototype that works really well with impressive accuracy. However, I noticed the signs of fatigue and self-doubt that often mean a project is abandoned at 80% without seeing the light of day.
So I decided to find a domain name and put the totally half-baked and not-yet-functioning version online. A crappy, embarrassing, non-functional version. But at least I might get some feedback, which is more likely to inspire me to keep going and work out how I could market this.
https://www.statementsamurai.com
I'm giving myself a week or two to refresh, then will press ahead some more and put the minimal functionality online while thinking about how to find users.
Yeah, about that lean thing...
Done
Hey, nice app. Can you add support for non-English transcription/translation, though?
Everything is done with real time calls to the eBay API. eBay's catalogue is too large to make it realistic to fetch it as a whole and, anyway, as far as I remember they don't offer a data dump (it's been a long time since I checked).
Luckily, eBay's API is very fast, but I also worked hard to keep the total response time of my code within a reasonable range.
Lumen (from Laravel) with Nginx on Linode.
PHP 7.1 has given a huge boost in terms of the performance you can get from a cheap VPS.
I block all bots and crawlers except Googlebot and Bingbot (the others bring no users but can hit the servers hard), so I'd treat Alexa etc with a pinch of salt.
Nonetheless, the figures I gave are for 2016. Since then, there has indeed been a fall in all numbers, largely because a site like this will always be overly reliant on Google and, for whatever reason, Google favours the site less this year than it did the year before.
At its peak, the site had approximately 53,000 daily visitors.
In my case it was word of mouth, social sharing, and the good graces of Google. Each individual user is simply worth too little to make a paid marketing strategy viable.
With the traffic I had, I did a lot of A/B testing to maximise conversions.
I built and run StoreSlider[1]. It made ~$700,000 in 2016, mainly in affiliate revenue from eBay. Costs are essentially hosting (between two and five $10 Linodes, depending on load).
Took me some effort to built, but it's on autopilot now.
I found myself spending far too much time on my iPhone, and most of this time was spent using Safari and Twitter for mindless browsing.
My approach to this was to uninstall all "problematic" apps, especially Twitter. Then I used my iPhone's Restrictions settings to hide Safari, and to prevent the installation of new apps. I then asked a friend to set a passcode on "Restrictions" so that I was unable to reinstall Safari, Twitter or anything else even if I wanted to.
After a few weeks, I got hold of the passcode again (to install something "essential" :)) but I have never since reinstalled Safari or Twitter. The habit was broken. Result: no longer do I walk down the street browsing the news, nor do I find myself staying up too late browsing in bed.
It's been a positive change for me.
I got bored of trawling eBay so decided to make my own tool to make it faster. StoreSlider has been going for a few years now and has been really useful.
Ha, I actually thought the logo looks similar to a cheap hotel/motel chain I saw when I was in the States, can't remember its name but it's not a good association :)
Sadly this is the limit of my photoshop skills for the time being.
Will update my JS to close the overlay on a zoom event, thanks for the tip.
Hi Imran,
I think you're right that the value prop isn't made explicit enough at the moment - I could use adding some kind of introductory text to make it clear what the site is about, question is where to put it.
To me, the value proposition is that 1. it shows you the absolute cheapest dates to go somewhere if you have the flexibility and 2. that it shows great deals to places you might never have thought of otherwise - all much easier to spot because the results are mapped.
Good suggestion re. target audience. Assuming you're talking about changes to the site itself (rather than its advertising) I'd be interested to hear what changes or additions you would make to go about that, i.e. to specifically target the 20-30 singles demographic.
Thanks for the advice, will look at options for being more specific on dates - although I suppose what I am really trying to offer here is an opportunity to see the absolute best prices on packages for those who have flexibility around their travel dates.
Thanks again!
Thanks for the really insightful comments. UI-wise, I can probably fix most of these (e.g. dates).
As you say, the real challenge is getting the traffic. Stumbleupon users are far from my target, but at least a cheap way to see how people interact with the site. Google Adwords is most likely best for targeting ready buyers, but without an idea of the expected value of each customer, may be too expensive at this point.
Thanks again for taking the time to comment in such depth, appreciated.
Yes, I should have disabled Heathrow for now, as no tour operator flies from there. Try any other airport and there are literally thousands of results.
Data comes direct from the tour operators.
EDIT: have now disabled Heathrow until I add data for that airport.