API QUERY We use the new

languages

(plural) attribute for deepgram latest model, see below. I want to hint at French or English.

curl --location --request PATCH "https://api.vapi.ai/assistant/4844b8ad-e4dd-45d3-b319-a9667567fdbb" --header "Content-Type: application/json" --header "Accept: application/json" --header "Authorization: Bearer XXX" --data "{
    \"transcriber\": {
        \"provider\": \"deepgram\",
        \"model\": \"flux-general-multi\",
        \"languages\": [
            \"fr\",
            \"en\"
        ],
        \"fallbackPlan\": {
            \"transcribers\": [
                {
                    \"model\": \"gemini-2.0-flash\",
                    \"language\": \"French\",
                    \"provider\": \"google\"
                }
            ]
        }
    }
}"

We get this error :

{
  "message": [
    "transcriber.property languages should not exist"
  ],
  "error": "Bad Request",
  "statusCode": 400
}

Sounds weird as the languages attribute is set in the swagger api definition : https://api.vapi.ai/api

Hey! I've dealt with this kind of schema mismatch before the Swagger docs and actual API validation are sometimes out of sync on Vapi's end. Quick thing to try: swap

languages

(plural) for

language

(singular) with just one value, e.g.

"language": "fr"

. The validator might not support the multi-language array yet despite it being in the docs. Happy to jump in and debug this with you if you need a hand! 🙋 @Jeebs from eva.be

Appreciate the help, except when coming from a bot AND not on point 🤣 The whole point of flux-general-multi is to allow multiple languages. Your solution to have only one is not helping much

Hey Jeebs, apologies for the delayed response. Are you still experiencing the issue? If so, please let me know. I'd be happy to help you out. Thanks, Akash