@Sangy It appears that VAPI currently does not support the

hi-Latn

language code for the Deepgram

nova-2

model. The error message indicates that the

transcriber.language

parameter must be one of the specified values, and

hi-Latn

is not listed among the supported language codes for the

nova-2-general

model within VAPI. Deepgram itself may support

hi-Latn

as a language code, but VAPI has its own set of supported languages and configurations. It's possible that VAPI has not yet updated its system to include

hi-Latn

for the

nova-2

model, or it may be a deliberate choice based on the services they offer. If you require

hi-Latn

to be configured with the Deepgram

nova-2

model through VAPI, I would recommend reaching out to VAPI support to inquire about the possibility of adding this language support or to get more information on when it might become available. Please ask support for more info on this.

Got it. Tagging @User for additional help.

using the API: {"message":["transcriber.language must be one of the following values for nova-2-general model: bg, ca, zh, zh-CN, zh-Hans, zh-TW, zh-Hant, cs, da, da-DK, nl, en, en-US, en-AU, en-GB, en-NZ, en-IN, et, fi, nl-BE, fr, fr-CA, de, de-CH, el, hi, hu, id, it, ja, ko, ko-KR, lv, lt, ms, multi, no, pl, pt, pt-BR, ro, ru, sk, es, es-419, sv, sv-SE, th, th-TH, tr, uk, vi","voice.provider must be one of the following values: 11labs, azure, cartesia, deepgram, lmnt, neets, openai, playht, rime-ai"],"error":"Bad Request","statusCode":400}

I advice to raise feature request.