HackerTube
HomeLongsLikedHistoryWatch laterCategoriesPlaylistsSettings

How the US Postal Service reads terrible handwriting

Tom Scott4 HN recommenders · recommended in 3 months, 2022-2026
YouTube
7M views · 4 years ago · documentary · evergreen

Tom Scott visits the USPS Remote Encoding Center in Salt Lake City, where keyers decipher mail images machines cannot read. It shows the lengths taken to decode bad handwriting.

Recommended in address and OCR threads as a look at the last human mail-coding operation.

Computer vision, speech & NLPManufacturing & industrial processesDocumentaries & storytelling#usps#tomscott#ocr#mailsorting#handwriting

8 comments from Hacker News

@shagie2026 · in “AI helps add 10k more photos to OldNYC”

As a profession (and under time constraints) ... Tom Scott : How the US Postal Service reads terrible handwriting - ▶ this video Part of the story is that the OCR that is handling hand lettered addresses. I also chuckled at the cursive letter recognition sheet on the side of the cube.

CitesView on HN
@yencabulator2026 · in “Using an Engineering Notebook”

> Though USPS has been doing handwriting recognition since the 80's. On a very restricted dictionary of possible content, with a pool of people employed to fix anything where the computer is unsure. ▶ this video

CitesView on HN
@Caligatio2024 · in “Postzegelcode”

And like almost all things that are interesting, here's a Ton Scott video about the same topic: ▶ this video

RecommendsView on HN
@LordDragonfang2024 · in “Amazon ditches 'just walk out' checkouts at its grocery stores”

Tom Scott did a great video showing the last one in operation. They use totally unique keyboards with what is essentially a stenography system: ▶ this video

RecommendsView on HN
@simmonmt2023 · in “What Is the Most Minimal UK Address Possible?”

This Tom Scott video shows some of the surprising (to me) lengths to which the USPS does go to figure out the right address: ▶ this video Obviously this doesn't apply to your case, but I certainly didn't think they'd have a national center for decoding terrible handwriting.

RecommendsView on HN
@bombcar2023 · in “What Is the Most Minimal UK Address Possible?”

▶ this video might be of interest - it has some of what the machines are trying to read and what accuracy they need.

RecommendsView on HN
@bombcar2023 · in “What Is the Most Minimal UK Address Possible?”

One of Don Rosa's books has the letters he sent to a friend in college, each with a drawing on the envelope and less and less address, until finally it was all drawing and barely part of a name - they all got (eventually) delivered. There's a specific branch of the post office that works on these - the dead letter office: ▶ this video

CitesView on HN
@omginternets2022 · in “Apple Kills Plans to Scan for CSAM in iCloud”

Actually, we're talking about both, and you are wrong on both counts. If you weren't, you would be able to point to an existing system that has demonstrated the capability of classifying an image set on the order of iCloud's, without producing a false-positive. You can't, because such a system doesn't exist. Even the "solved" problem of OCR isn't capable of such a feat: ▶ this video

CitesView on HN
I sent robot forgeries to a handwriting expert23:39

I sent robot forgeries to a handwriting expert

Stuff Made Here
11.5M views • 3 years ago
The Worst Typo I Ever Made11:25

The Worst Typo I Ever Made

Tom Scott
6.4M views • 6 years ago
Breaking The Creepy AI in Police Cameras35:49

Breaking The Creepy AI in Police Cameras

Benn Jordan
3.8M views • 1 year ago
BFS-Auto: High Speed Book Scanner at over 250 pages/min1:00

BFS-Auto: High Speed Book Scanner at over 250 pages/min

Ishikawa Group Laboratory
1.3M views • 13 years ago
why typing like this is sometimes okay.4:31

why typing like this is sometimes okay.

Tom Scott
5.4M views • 7 years ago
Uppestcase and Lowestcase Letters  [advances in derp learning]24:13

Uppestcase and Lowestcase Letters [advances in derp learning]

suckerpinch
477K views • 5 years ago
AI Language Models & Transformers - Computerphile20:39

AI Language Models & Transformers - Computerphile

Computerphile
354K views • 7 years ago
Why Electronic Voting Is Still A Bad Idea12:01

Why Electronic Voting Is Still A Bad Idea

Tom Scott
5.1M views • 6 years ago
Why Electronic Voting is a BAD Idea - Computerphile8:21

Why Electronic Voting is a BAD Idea - Computerphile

Computerphile
2.3M views • 11 years ago
I tried using AI. It scared me.15:49

I tried using AI. It scared me.

Tom Scott
8.5M views • 3 years ago
Stealing Baseball Signs with a Phone (Machine Learning)13:30

Stealing Baseball Signs with a Phone (Machine Learning)

Mark Rober
30M views • 7 years ago
Let's build the GPT Tokenizer2:13:35

Let's build the GPT Tokenizer

Andrej Karpathy
1.3M views • 2 years ago
Why Web Filters Don't Work: Penistone and the Scunthorpe Problem3:37

Why Web Filters Don't Work: Penistone and the Scunthorpe Problem

Tom Scott
3.6M views • 10 years ago
Ch(e)at GPT? - Computerphile13:52

Ch(e)at GPT? - Computerphile

Computerphile
617K views • 3 years ago
The Artificial Intelligence That Deleted A Century6:16

The Artificial Intelligence That Deleted A Century

Tom Scott
2M views • 7 years ago
Deadly Truth of General AI? - Computerphile8:30

Deadly Truth of General AI? - Computerphile

Computerphile
941K views • 11 years ago
The Art of the Bodge: How I Made The Emoji Keyboard17:15

The Art of the Bodge: How I Made The Emoji Keyboard

Tom Scott
3.5M views • 11 years ago
Windows Vista Speech Recognition Tested - Perl Scripting10:34

Windows Vista Speech Recognition Tested - Perl Scripting

scrubadub1
913K views • 19 years ago
Glitch Tokens - Computerphile19:29

Glitch Tokens - Computerphile

Computerphile
353K views • 3 years ago
LSTM is dead. Long Live Transformers!28:48

LSTM is dead. Long Live Transformers!

Seattle Applied Deep Learning
540K views • 6 years ago