Spanish AI voiceover

The Spanish voices, with samples

Each sample reads the same line so you can compare them fairly.

Castilian by default, Latin American by search

Spanish resolves to the es-ES locale, so the automatic voice is from Spain, with the distinción a Mexican or Colombian audience will notice immediately. That default is fine for a European market and wrong for a Miami one. The fix takes a minute: open the voice dialog on a scene and search the live catalogue, which carries es-US among other Spanish locales, then apply the same voice scene by scene. The direction of the writing does not change either way, because the script itself is neutral unless your brief asks for regional vocabulary, coche versus carro being the classic case worth deciding in the brief.

How the eight voices divide the work

The energetic pair, Puck and Laomedeia, carry retail promos and social ads. Charon and Erinome are the instructional two, and any tutorial gets them regardless of what the plan requested, a deliberate rule so a how-to never arrives sounding like a commercial. Umbriel and Sulafat suit a restaurant opening or a testimonial where warmth sells. Iapetus and Aoede handle corporate updates. The director chooses style and gender from your brief's tone, and the choice binds the whole video until you override a scene.

Spanish runs long, and the pacing knows it

The same idea takes roughly a fifth more syllables in Spanish than in English, so a line that fits a four second English scene can crowd its Spanish twin. Scenes stretch automatically to fit their audio, and in the dynamic pace a scene grows at most one and a half times its planned length while the rest of the line flows over the next cut. If a scene still feels rushed, the per-scene speed control drops to 0.85x, or you shorten the line, which is usually the better edit. A delay in seconds on each scene lets the visual land before the voice enters.

Numbers, names and the written word

The engine reads what is written, so decide how prices should be spoken and write them that way: diecinueve noventa is unambiguous where 19,90 invites a robotic decimal read. Foreign names get respelled until they sound right, there being no phonetic input. Punctuation is load bearing, since sentence breaks become caption cues and question marks lift the intonation. Keep the inverted openers, ¿ and ¡, because the voice uses the whole sentence shape. The director is barred from em dashes in narration, so a hand-pasted one is the only way they appear, and they break the rhythm the captions expect.

Captions in this script

Spanish captions behave exactly like English ones: each word lights up as it is spoken, timed against the real clip, inside a fixed dark pill that keeps white text legible over any scene. Accented vowels and the ñ are part of the basic Latin extension that every font in the catalogue ships, so no look breaks them. The one thing to watch is line length: Spanish sentences run longer than their English equivalents, and the pill wraps a sentence over 48 characters into a second cue, which reads fine but doubles the caption changes on a fast scene.

A video narrated in Spanish

Restaurant promo video example

24s promo

Questions people ask

Related pages