# Re: Shirt Reader (#140)

**URL:** <https://www.ruby-forum.com/t/re-shirt-reader-140/115305>\
**Category:** Ruby\
**Created:** [September 23, 2007, 2:44pm UTC](https://www.ruby-forum.com/t/re-shirt-reader-140/115305 "2007-09-23T14:44:25Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![steve\_d](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/s/ce73a5/32.png) [@steve\_d](https://www.ruby-forum.com/u/steve_d)\
**Post date:** [September 23, 2007, 2:44pm UTC](https://www.ruby-forum.com/t/re-shirt-reader-140/115305/1 "2007-09-23T14:44:25Z")

</div>

Hi all,

my solution: [http://rn86.net/~stevedp/tshirt\_reader.tar.gz](http://rn86.net/~stevedp/tshirt_reader.tar.gz)

For my solution I used a combination of the Metaphone algorithm,  
pronunciation matching (via CMU pronunciation dictionary  
[[The CMU Pronouncing Dictionary](http://www.speech.cs.cmu.edu/cgi-bin/cmudict)]),  
and the Levenshtein distance algorithm. The input must be words or  
numbers which sound out the answer word. It will give back at max 10  
words that are possible matches. In most of my test words the correct  
match is in the 1st or 2nd place, but a few are in 5th or more.

You must first run prepare\_dicts.rb which does some preparation work.

You  
need the Text gem installed (gem install text) for the  
Metaphone/Levenshtein algorithm. I used ruby inline to re-implement  
the Levenshtein algorithm in C (versus the Text gem’s pure ruby impl.)  
which made it run like 20x faster at least. If you don’t have ruby  
inline installed it will fall back on the Text gem.

Here’s the final list of phrases I was testing with (taken from  
test/test\_tshirt.rb)

```
%w[e scent shells] => 'essentials',
%w[q all if i] => 'qualify',
%w[fan task tick] => 'fantastic',
%w[b you tea full] => 'beautiful',
%w[fun duh mint all] => 'fundamental',
%w[s cape] => 'escape',
%w[pan z] => 'pansy',
%w[n gauge] => 'engage',
%w[cap tin] => 'captain',
%w[g rate full] => 'grateful',
%w[re late shun ship] => 'relationship',
%w[con grad yeul 8] => 'congratulate',
%w[2 burr q low sis] => 'tuberculosis',
%w[my crows cope] => 'microscope',
%w[add minus ray shun] => 'administration',
%w[accent you ate it] => 'accentuated',
%w[add van sing] => 'advancing',
%w[car knee for us] => 'carnivorous',
%w[soup or seed] => 'supercede',
%w[poor 2 bell o] => 'portobello',
%w[d pen dance] => 'dependence',
%w[s o tear rick] => 'esoteric',
%w[4 2 it us] => 'fortuitous',
%w[4 2 n 8] => 'fortunate',
%w[4 in R] => 'foreigner',
%w[naan disk clothes your] => 'nondisclosure',
%w[Granmda Atika Lee] => 'grammatically',
%w[a brie vie a shun] => 'abbreviation',
%w[pheemeeneeneetee] => 'femininity',
%w[me c c p] => 'mississippi',
%w[art fork] => 'aardvark',
%w[liberty giblet] => 'flibbertigibbet',
%w[zoo key knee] => 'zucchini',
%w[you'll tight] => 'yuletide',
%w[Luke I like] => 'lookalike',
%w[mah deux mah zeal] => 'mademoiselle',
%w[may gel omen yak] => 'megalomaniac',
%w[half tell mall eau gist] => 'ophthalmologist',
%w[whore tea cull your wrist] => 'horticulturist',
%w[pant oh my m] => 'pantomime',
%w[tear a ball] => 'terrible',
%w[a bowl i shun] => 'abolition',
%w[pre chair] => 'preacher',
%w[10 s] => 'tennis',
%w[e z] => 'easy',
%w[1 door full] => 'wonderful',
%w[a door] => 'adore',
%w[hole e] => 'holy',
%w[grand your] => 'grandeur',
%w[4 2 5] => 'fortify',
%w[age, it ate her] => 'agitator',
%w[tear it or eel] => 'territorial',
%w[s 1] => 'swan'

```

- steve

---

<div class="post-metadata">

**Author:** ![steve\_d](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/s/ce73a5/32.png) [@steve\_d](https://www.ruby-forum.com/u/steve_d)\
**Post date:** [September 23, 2007, 8:10pm UTC](https://www.ruby-forum.com/t/re-shirt-reader-140/115305/2 "2007-09-23T20:10:53Z")

</div>

On Sun, 23 Sep 2007 21:42:30 +0900, steve d wrote:

> 2nd place, but a few are in 5th or more.  
> My answer’s along the same lines with Metaphone, but nowhere near as  
> good  
> as steve’s:

require ‘rubygems’  
require ‘text’  
include Text::Metaphone

#use this to do the double\_metaphone as a drop-in replacement for  
metaphone  
def dmetaphone word  
first,second = double\_metaphone word  
second || first  
end

#this solution gets 3 of the test cases correct if single metaphone is  
used  
#it gets 10 of the test cases correct if double-metaphone is used, but  
also  
#provides a much longer list of wrong answers for everything

# 

#use this alias to set the particular phonetic conversion algorithm  
alias\_method :phonetic\_convert, :dmetaphone

NUMBERS=Hash.new{|h,k| k}.merge!({“1”=\>“one”, “2”=\>“two”, “3”=\>“three”,  
“4”=\>“four”,“5”=\>“five”,“6”=\>“six”,“7”=\>“seven”,“8”=\>“eight”,“9”=\>“nine”})

DICT=open(‘/usr/share/dict/words’) do |f|  
d=Hash.new{|h,k| h[k]=[]}  
f.each\_line do |word|  
word=word.chomp  
d[phonetic\_convert(word).gsub(/\s/,‘’)] \<\< word  
end  
d  
end

def rebus words  
words=words.collect{|x| NUMBERS[x]}.join(’ ‘)  
DICT[phonetic\_convert(words).gsub(/\s/,’')]  
end

#tests given by steve d [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid)  
expectations = {  
%w[e scent shells] =\> ‘essentials’,  
%w[q all if i] =\> ‘qualify’,  
%w[fan task tick] =\> ‘fantastic’,  
%w[b you tea full] =\> ‘beautiful’,  
%w[fun duh mint all] =\> ‘fundamental’,  
%w[s cape] =\> ‘escape’,  
%w[pan z] =\> ‘pansy’,  
%w[n gauge] =\> ‘engage’,  
%w[cap tin] =\> ‘captain’,  
%w[g rate full] =\> ‘grateful’,  
%w[re late shun ship] =\> ‘relationship’,  
%w[con grad yeul 8] =\> ‘congratulate’,  
%w[con grad yule 8 shins] =\> ‘congratulations’, #from Phrogz  
%w[2 burr q low sis] =\> ‘tuberculosis’,  
}

expectations.each do |words,target|  
result=rebus(words)  
if result.include?(target)  
printf “%s correctly gave %s.\n”, words.inspect, target  
else  
printf “%s incorrect. Expected %s.\n”, words.inspect, target  
end  
printf “Metaphone of words: %s Metaphone of target: %s\n”,  
phonetic\_convert(words.collect{|x| NUMBERS[x]}.join(’ ')),  
phonetic\_convert(target)  
printf “Matching words %s\n”, result.inspect  
end

---

<div class="post-metadata">

**Author:** ![steve\_d](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/s/ce73a5/32.png) [@steve\_d](https://www.ruby-forum.com/u/steve_d)\
**Post date:** [September 25, 2007, 9:46pm UTC](https://www.ruby-forum.com/t/re-shirt-reader-140/115305/3 "2007-09-25T21:46:43Z")

</div>

I’m wondering if this quiz would work in the opposite direction  
reasonably well? Given a word (or a sentence), create a series of  
words that will sound it out. If you stick to using nouns as the  
sound words, you can pretty reliably get an image for the word using  
Google’s picture search.

Cheers,  
B