Word to Number Converter

Convert English number words to digits: "forty-two" → 42.

Word numbers (one per line)

or drop a file here

Digits

forty-two→42

one hundred and twenty-three→123

three thousand five hundred→3,500

Supported word number formats

This converter handles standard English number words including ones (zero–nineteen), tens (twenty–ninety), and multipliers (hundred, thousand, million, billion, trillion). Hyphens and "and" are treated as separators: "twenty-four" and "twenty four" both work. Use "negative" or "minus" as a prefix for negative numbers. Commas in numbers like "one,000" are ignored. Related: number to ordinal converter and number base converter.

Built and maintained by Meet Shah · Last updated 28 July 2026

What this tool is used for

Parsing a written amount from a scanned cheque or a contract into a numeral.
Normalising quantities written as words in a spreadsheet import.
Checking that a written number and its numeral in a document agree.
Converting spoken-style input captured from a transcript.
Producing numerals from a form where users typed words.

Frequently Asked Questions

What is the difference between short and long scale?: A genuine international divide. In the short scale (US, modern UK) a billion is 10^9; in the long scale (much of Europe) it is 10^12 and 10^9 is a milliard. The same word means numbers a thousand times apart.
Is 'and' allowed in the words?: British English uses it — "one hundred and five" — while American English omits it. Parsers must accept both, and the word carries no value, so it is discarded rather than treated as an operator.
How are hyphens handled?: Compound numbers from twenty-one to ninety-nine are conventionally hyphenated. A parser must treat the hyphen as a separator, or "twenty-one" fails to match either component word.
Are ordinals supported?: They need separate handling — "twenty-first" is 21st, not 21. Some are irregular (first, second, third, fifth, ninth, twelfth), which is why a simple suffix rule produces wrong results.
What is this useful for?: Parsing amounts written out on cheques and contracts, normalising voice-transcription output where numbers arrive as words, and processing legal or financial documents where figures are spelled for unambiguity.
How are amounts written on cheques parsed?: With their own conventions — "and" separating the fractional part, and often a trailing "only". They are a legal formulation rather than plain arithmetic, which is why a general parser needs the special case.
How does the Indian numbering system differ?: It uses lakh (100,000) and crore (10,000,000), which have no short-scale equivalent. "Two crore" is 20,000,000, and a parser that only knows million and billion cannot express it at all.

Common errors and gotchas

Mixing the short and long scale, where a billion means different things in different conventions.
Overlooking the 'and' in British usage, which is grammatical rather than arithmetic.
Assuming hyphenation is consistent, since forty-two and forty two both appear.
Losing an ordinal, where 'third' is a position and not the number three in every context.
Expecting fractional words like 'a half' to parse, which many implementations do not handle.