Languages Vidnoz publishes for

Vidnoz (vidnoz.com) declares 12 language alternates and serves 11 of them from language subdomains, so one is counted here. Its talking-photo page carries two different spoken language counts, 100+ and 140+, within the same page. As of 2026-09-22.

Inclusion rule. The same three columns are kept for every tool in the register; a column the vendor does not publish is recorded as no list rather than filled from a product description. Order. Fixed column order, identical on every tool page.

1Two counts on one page

The talking-photo page offers, within the same document, a promise to create a talking photo in 100+ languages and a promise that the photo will speak emotively and naturally in 140+ languages, with a heading repeating the larger figure. Forty languages separate the two, and nothing on the page distinguishes what the smaller number covers that the larger one does not.

There is a benign reading available: one figure may count languages and the other languages plus accents, or one may belong to a free tier. The page does not say, so the register does not guess. Both are printed in the cell, which is the only treatment that is honest about what a visitor is actually given, and neither is turned into a list because neither has one.

2Twelve alternates, eleven of them on other hosts

The declared set is Arabic, German, English, Spanish, French, Italian, Japanese, Korean, Dutch, Portuguese, Turkish and traditional Chinese. Eleven of the twelve resolve to language subdomains reached by their own names, leaving the English one on the host that was read. Under the rule used for every vendor here, only same-host alternates are counted, so this row is one.

Rask AI is in the same position with fifteen of sixteen, and between them the two rows show why the rule is worth having even though it looks harsh. Counting other hosts would make a vendor that runs twelve properties look twelve times as localised as a vendor that translated its own pages, and the column would stop being a measurement. The unadjusted figure is printed here so a reader can disagree with the specific step rather than with the table.

3What can be checked before paying, and what cannot

A reader can check the declaration, because it is in the markup, and can check that the subdomains exist. Those are the facts the site column carries and they are small ones. A reader cannot check whether Turkish is among the 140, or the 100, or only one of them, because no set is published in either place.

That is the same gap four vendors in this register leave, and it is worth naming precisely. A count tells a buyer how large a vendor believes its inventory to be. A list tells a buyer whether their own market is in it. Only the second question has money attached to it, and on this row it is the one that has no answer on any page a visitor can open.

4Sources

Alternates counted from the markup of vidnoz.com; the two counts are quoted from its talking photo page, both read 2026-09-22.

5Where this entry sits in the register

Inclusion rule. Tools with a public English site selling AI video generation or AI video dubbing. Site languages count hreflang alternates served from the host being read; x-default and alternates pointing at another host are not counted, for every vendor alike. Order. Alphabetical by tool name.

Next entry in the register: CapCut. Previous entry: VEED. Every tool side by side is in the register, and the rule behind each count is on the how counted page.

• Site languages
  12 alternates declared: ar, de, en, es, fr, it, ja, ko, nl, pt, tr, zh-Hantraw count before the rule is appliedvidnoz.com / recorded 2026-09-22
• Site languages
  Eleven of the twelve are served from language subdomains; only the English alternate points at the host being countedwhere the alternates resolvevidnoz.com / recorded 2026-09-22
• Voice languages
  One page carries both “in 100+ languages” and “speak emotively and naturally in 140+ languages”two counts on one pageVidnoz, talking photo page / recorded 2026-09-22