Convert a numeric scalar to a string with the number value given in English words: for example the value 1024 is returned as 'one thousand and twenty-four'.
NUM2WORDS accepts a scalar of any standard numeric class: single, double, int, or uint. Integer types are parsed with their full precision, while floating types have internal precision limits to ensure least-unexpected output.
Options allows the user to select:
* the number type selection: ordinal / decimal / cheque / money / highest multiplier.
* the number of significant digits or order of precision.
* upper / lower / title / sentence case.
* trailing zeros or without.
* the use of a comma between magnitude groups.
* the use of a hyphen between tens and ones.
* the use of 'and' before the tens/ones (required in British and Commonwealth English).
* a 'positive' prefix for values >=+0.
* the numbering scale: short / long / Indian / Peletier / Rowlett / Knuth (-yllion).
* currency unit names (Pound, Pence, Dollar, Cents, Rupees, etc).
### Bonus Functions / Scripts ###
Three bonus files are also included:
* NUM2WORDS_DEMO compares the output of NUM2WORDS with real-world example numbers.
* NUM2WORDS_FAST is a minimalist version without any options, but twice as fast.
* NUM2WORDS_RAT converts a numeric to a string with an improper fraction in words.
### Convert Words to Numbers ###
### Examples ###
ans = 'zero'
ans = 'one thousand and twenty-four'
ans = 'negative one thousand and twenty-four'
num2words(1024, 'pos',true, 'case','title', 'hyphen',false)
ans = 'Positive One Thousand and Twenty Four'
num2words(1024, struct('type','ordinal', 'case','sentence'))
ans = 'One thousand and twenty-fourth'
num2words(1024, 'and',false, 'order',1) % round to the tens.
ans = 'one thousand twenty'
num2words(pi, 'order',-10) % round to tenth decimal digit
ans = 'three point one four one five nine two six five three six'
num2words(intmax('uint64'), 'sigfig',3, 'comma',false)
ans = 'eighteen quintillion four hundred quadrillion'
num2words(intmax('uint64'), 'sigfig',3, 'type','highest')
ans = 'eighteen point four quintillion'
num2words(intmax('uint64'), 'sigfig',3, 'scale','long')
ans = 'eighteen trillion, four hundred thousand billion'
num2words(intmax('uint64'), 'sigfig',3, 'case','title', 'scale','indian')
ans = 'One Lakh, Eighty-Four Thousand Crore Crore'
num2words(intmax('uint64'), 'order',17, 'case','upper', 'scale','knuth')
ans = 'EIGHTEEN HUNDRED FORTY BYLLION'
num2words(1234.56, 'type','cheque', 'unit','Euro')
ans = 'one thousand, two hundred and thirty-four euro and fifty-six cents'
num2words(1234.56, 'type','cheque', 'unit','Pound|', 'subunit','Penny|Pence')
ans = 'one thousand, two hundred and thirty-four pounds and fifty-six pence'
num2words(101, 'type','money', 'unit','Dalmatian|', 'case','title')
num2words(1001, 'type','money', 'unit','Night|', 'case','title')
[num2words(2e4, 'type','money', 'unit','League|', 'case','title'), ' Under the Sea']
Oh well, I guess my promise of a sixth star was an empty one, with a 5 star limit on the FEX. :)
Well done anyway. This does everything I can think of in the way of converting numbers to a textual form.
Based on the comment from John D'Errico, now supports money/cheque number types.
John D'Errico: Thank you for your feedback.
A money 'type' would be interesting, but limiting it to dollars/cents would definitely start a culture-war on FEX, and the subunit handling is challenging: some currencies use a ratio other than 1/100 to define the subunit, a few currencies don't even have subunits, and the Yen has two subunits. So it would require an option for the user to supply their currency name/s, some complicated subunit handling, and a way to decide whether to use 'only' or not (value:no, cheque:yes). As the 'money' option is basically just repeated applications of the function, I figured an example would suffice.
Lots of options to control the style of your output. Good help. Well done.
I was surprised not to see a direct money option though. So a 'type' of 'dollars' might have been useful and interesting. I'd have had to figure out how to give it an extra star then.
* Check for cell array optional inputs.
* Include HTML _DOC file in zip.
* Change HTML documentation filename.
* Add HTML directory
* Rename "myriad" scale to "knuth".
* Improve documentation.
* Update license.
* Add 'india' number scale (lakh and crore).
* Simplify cell array creation.
* Fix bug in myriad parsing (if >1 highest level multiplier).
* Add a new <scale> option <myriad> (Knuth's -yllion).
* More robust handling of some edge cases: zero, NaN, Inf.
- Minor edit in blurb.
- Minor edit to FEX blurb.
* Add money/cheque number type.
- I accidentally uploaded an older version. This version matches the online description :)
* comma control option.
- Downloadable as toolbox
- Fix bug for ordinals 'first' and 'second'.
- Improve decimal fraction handling.
Inspired by: text2speech - tts, num2english, text-to-speech, Variable Precision Integer Arithmetic, Number to Scientific Prefix, Text2Speech for Matlab using unofficial google service, Date Vector/Number to ISO 8601 Date String, Customizable Natural-Order Sort, ISO 8601 Date String to Serial Date Number, Numeric to Ordinal-String, Natural-Order Row Sort, Natural-Order Filename Sort, Convert number to words, Number to Myriad, Words to Number
Inspired: Words to Number, Round to Electronic Component Values, Scientific Prefix to Number, Number to Myriad, Number to Scientific Prefix, Customizable Natural-Order Sort, Numeric to Ordinal-String, Natural-Order Row Sort, Natural-Order Filename Sort, Interactive Regular Expression Tool