A Latin-script alphabet (Latin alphabet or Roman alphabet) is an alphabet that uses letters of the Latin script. The 21-letter archaic Latin alphabet and the 23-letter classical Latin alphabet belong to the oldest of this group.^[1] The 26-letter modern Latin alphabet is the newest of this group.

This article needs additional citations for verification. (January 2021)

The 26-letter ISO basic Latin alphabet (adopted from the earlier ASCII) contains the 26 letters of the English alphabet. To handle the many other alphabets also derived from the classical Latin one, ISO and other telecommunications groups "extended" the ISO basic Latin multiple times in the late 20th century. More recent international standards (e.g. Unicode) include those that achieved ISO adoption.

Apart from alphabets for modern spoken languages, there exist phonetic alphabets and spelling alphabets in use derived from Latin script letters. Historical languages may also have used (or are now studied using) alphabets that are derived but still distinct from those of classical Latin and their modern forms (if any).

The Latin script was typically slightly altered to function as an alphabet for each different language (or other use), although the main letters are largely the same. A few general classes of alteration cover many particular cases:

• diacritics could be added to existing letters;
• two letters could be fused together into ligatures;
• additional letters could be inserted; or
• pairs or triplets of letters could be treated as units (digraphs and trigraphs).

These often were given a place in the alphabet by defining an alphabetical order or collation sequence, which can vary between languages. Some of the results, especially from just adding diacritics, were not considered distinct letters for this purpose; for example, the French é and the German ö are not listed separately in their respective alphabet sequences. With some alphabets, some altered letters are considered distinct while others are not; for instance, in Spanish, ñ (which indicates a unique phoneme) is listed separately, while á, é, í, ó, ú, and ü (which do not; the first five of these indicate a nonstandard stress-accent placement, while the last forces the pronunciation of a normally-silent letter) are not. Digraphs in some languages may be separately included in the collation sequence (e.g. Hungarian CS, Welsh RH). New letters must be separately included unless collation is not practised.