D-ID: a language field with no list behind it
The reference has a language field. What it does not have is a list of what goes in it. The languages arrive with the five named speech providers, which means the answer lives on somebody else's documentation rather than this one. As of 2026-09-22.
| What the documentation settles | What it leaves to a take |
|---|---|
| An optional language field sits beside the chosen voice | Which values that field accepts, since none is listed |
| Five speech providers are named for the voice: Microsoft, ElevenLabs, Amazon, Google and Azure OpenAI | Whose language list applies once a provider has been chosen |
| The voice is a voice id selected from the list of available voices | How many of those voices cover a given market |
Inclusion rule. Only statements found on the vendor page named in the caption, and only ones bearing on this one field. Nothing here is filled in from generated output. Order. In the order the statements arrive on the vendor's page.
1A field that delegates its own contents
Naming the parameter and not its range is a particular kind of half-answer. It confirms that language is something a caller controls, which several entries here do not even manage, and then hands the enumeration to whichever speech provider was picked. The reference is accurate; it is simply not the document that answers the question.
For a production that means the language check is a check on a supplier one level down, and it has to be repeated for each provider under consideration. A list that lives in five places is harder to audit than a short list that lives in one.
2Delegation moves the risk as well as the list
Where the voices come from an outside catalogue, the coverage can change without the product changing. A provider deprecating a voice, or reorganising its language codes, reaches a series through a dependency the talk reference never mentions.
That is worth writing down beside the convenience. Five providers is a wide net and a wide surface, and nothing on this page tells a reader which of the five a particular voice id belongs to before the call is made.
3The others that claim languages without naming any
Two more entries here assert more than one language and name none of them. Each of the three does it differently: a delegated field, a claim of full support, and a list of performance types instead of places.
- Per-character bindingThe voice is a voice id selected from the list of available voices, with an optional language field beside ita voice id on every request
- Voice sourceFive speech providers are named for the voice: Microsoft, ElevenLabs, Amazon, Google and Azure OpenAIoutside catalogues named on the page
4Sources
Read from the create a talk reference at docs.d-id.com on 2026-09-22. The same column across every entry is on languages; everything this vendor publishes about speech is on D-ID. What counts as documented is on how read.