One endpoint to speak a phrase. Machine-readable files are provided for AI assistants.
Create a key in the dashboard, then request audio.
curl https://dev.speakvora.com/api/v1/speak \
-H "Authorization: Bearer $SPEAKVORA_KEY" \
-H "Content-Type: application/json" \
-d '{"text":"Aim assist is off.","voice":"af_heart","lang":"en"}' \
POST /v1/speak returns JSON with a url to the audio (m4a). Full schema in the OpenAPI spec.
Text that is not in the phrase library is voiced on demand for English Kokoro voices (af_heart and others) and returned as an mp3 url. First request for new text takes a few seconds; the same text afterwards returns instantly. Other languages and Premium voices use the phrase library for now.
Pass language as an ISO code or auto. Ambiguous text is rejected rather than guessed. Non-English clips carry native_review_pending until a native speaker approves them.
Text you send and the audio produced for it are stored to improve speed and quality. Never send personal data: use placeholders such as (name) and (date).
Point your coding assistant at /docs/llms.txt and the OpenAPI spec; they contain everything needed to wire up the API.