Speech (TTS)

Status: Live / on request. Same endpoint for every Lemka voice. English Kokoro on /speak is not this product.

POST https://<HOST>/v1/audio/speech/{language}

Authorization: Bearer <API_KEY>

Field	Type	Required	Description
`text`	string	Yes	Text in that language.
`format`	string	No	Default `wav`.

{
  "language": "kha",
  "audio_b64": "<base64 wav>",
  "format": "wav",
  "request_id": "req_8f2a1c"
}

{language} is one of kha, grt, trp, lus, nag, mni. A code not on your image returns 503 model_unavailable.