Hyphenate (Spanish)

This script divides Spanish words into syllables.

Since Spanish pronunciation follows ortography, it would be expected that syllabification based on ortography only would be possible. This is almost true. The problems come with diphtongs and hiatuses, since in different dialects these are made in different situations. Orthographical accents (“tilde” in Spanish) are supposed to help disambiguating pitch accent, and thus determining whether two vowels do or not form a diphtong. However, many of these rules rely on the speaker’s knowledge of the prosody of a given word, so can’t be used effectively only with orthography.

When possible, the rules stated by the Real Academia de la Lengua (supervisor body of Spanish) are followed.

Consonants

Consonants can be divided in a number of groups depending on the combinations with other consonants they allow, and their order in this clusters. The groups’ names are probably very linguistically incorrect, but the author is not a linguist and very lacking in proper terminology.

In the special digraphs qU and gU, the uppercase U is used to indicate that this is not a vowel, so it is not erroneously incorporated into the nucleus.

Spanish allows much shorter consonant clusters than English, and onset and coda are clearly different.

Vowels

Vowels with a “tilde” (accent) have the pitch accent of the word, so they are always the strong partner in a diphtong. Some sources state that when a tilde is present, there is no diphtong, but the author has not found conclusive arguments, and in his particular dialect this happens to not (always) be true.

The nucleus is thus defined as a vowel, dipthong or tripthong. The special vowel ü is always part of a dipthong, since the dieresis is used to indicate that the u is to be pronounced before an e or i (usually silent after q or g). y at the end of word is pronounced as i.

Hyphenating

First, the syllable nucleus must be found, with a longest match approach since we want to allow dipthongs. When we find the nucleus, we mark it with NUC to both sides. Then syllables, defined as onset + nucleus + coda are found.

Two steps are needed because syllable matching is shortest match. This encodes the Maximal Onset principle (actually, the equivalent Minimal Coda one).

Hyphens are inserted after each syllable, provided it is followed by a valid syllable as context.

First u is marked in the digraphs in qui, que, gui and gue. Intervocalic y is realized as a consonant, so it is marked so as to not be put in a nucleus (e.g. ayer)