Seventy-two languages it can hear, twenty-nine it can speak
VEED publishes two lists on one support page: 72 languages it can detect in an uploaded file and 29 it can dub into. No other vendor here separates the languages a tool can hear from the ones it can produce. As of 2026-09-22.
| Side of the list | Entries | What it limits |
|---|---|---|
| Languages detectable in the source file | 72 | Which recordings can be used as input |
| Languages the tool can dub into | 29 | Which markets a finished video can reach |
| Difference | 43 | Languages it can hear but not speak |
| A total printed on either list | None | Both figures come from counting |
Inclusion rule. Both sides of one vendor's published list, counted from the entries, with the difference stated rather than explained. Order. Input side first, then output side, then the difference.
1The shorter number is the product
Forty-three languages can be recognised and not spoken. For a buyer the output list is the specification: a tool with 29 target languages sells into 29 markets however many it can transcribe.
Publishing both is more honest than publishing either alone. A vendor quoting only the larger figure would be describing its speech recognition while appearing to describe its dubbing, which is the same conflation this register splits the columns to avoid.
2Why the two sides differ at all
Recognising speech and synthesising it are separate models. Recognition can be trained from recordings of people talking, of which there is a great deal in many languages; synthesis needs a voice built and licensed for each one.
So the asymmetry is structural rather than a gap somebody forgot to close, and it should be expected at every dubbing vendor. This is the only one here whose pages let a reader see it.
3Two lists, no totals, and an empty site cell
Neither list states its own length; the figures here are counted from the entries. The same vendor declares no language alternates on its own host, so its site cell is empty and the support article holding both lists is served in English.
That combination, a detailed output list behind an untranslated shop window, appears three times in this register and always at a vendor selling dubbing. The language question, for that business, lives entirely in the file being delivered.
4Sources
Both lists counted from the VEED dubbing languages article. The column itself is described on dialogue language, and every cell VEED fills is on its tool page.
5The same column, tool by tool
| Tool | Voice languages |
|---|---|
| CapCut | No list published |
| D-ID | 120+, count only |
| Fliki | 91 named, plus 117 dialects |
| Hailuo | No list published |
| HeyGen | Two named lists, no total |
| Higgsfield | No list published |
| invideo AI | No list published |
| Kling AI | Five, count only |
| LTX Studio | No list published |
| Pika | No list published |
| PixVerse | No list published |
| Rask AI | 135+, count only |
| Runway | No list published |
| SceneMixer | 15, plus Cantonese for dialogue |
| Synthesia | 143 named |
| VEED | 29 dub-to, 72 detectable |
| Vidnoz | 100+ and 140+, counts only |
| Vidu | No list published |
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.
Neighbouring cells: Synthesia: dialogue language and Vidnoz: dialogue language. All of them together: the cell index and the register table.
- Voice languagesA list headed “Supported spoken languages:” with 72 entries, introduced as “the languages that we can detect in the original file”input side of the dubbing list
- Voice languagesA second list headed “Supported Target languages” with 29 entries, introduced as “the languages we can dub to”output side of the dubbing list
- Voice languagesNeither list carries a total; the two are printed one after the other on the same pagehow the lists are presented