STT (Speech-to-Text) converts what the player says into text for the voice matching system. The script ships with a built-in STT backend — no key is required to get started.
Adding a BYOK STT key unlocks:
Language = "auto" — auto-detect the spoken language per playerGroq is the recommended STT provider. It has a generous free tier and is the only provider that enables Language = "auto" (automatic language detection).
Create a free account at console.groq.com and generate an API key.
You can configure a second STT provider. If the primary provider fails, the system automatically falls back to the secondary. Both providers can be different (e.g. Groq primary, Azure secondary).
stt_provider_2 is required when stt_api_key_2 is set. If omitted, the secondary slot is ignored.
These control how voice input is translated before command matching. Set in server.cfg — not in config.lua.
| Convar | Default | Description |
|---|---|---|
voice_translate | "auto" | "auto" | "true" | "false" — enable/disable translation |
voice_translate_target | "en" | Language your command phrases are written in |
voice_dual_match | "false" | Match against both raw and translated text simultaneously |
voice_dual_match_priority | "highest_score" | "highest_score" | "raw_first" | "translated_first" |
"auto" enables translation only when Config.Language differs from voice_translate_target. voice_dual_match is useful when players speak in a non-English language but your phrases are in English — it matches against both the raw transcript and the translated version and picks the best result.
groq_api_key is deprecated. It still works as a fallback (for both STT and LLM) but will be removed in a future version. Migrate to stt_api_key + stt_provider "groq".
Last updated about 9 hours ago