From da89108b82a5c33d7972fd7bc248a379394d5a5c Mon Sep 17 00:00:00 2001 From: Peter Steinberger Date: Sat, 25 Apr 2026 04:53:33 +0100 Subject: [PATCH] docs(minimax): document tts env vars --- docs/providers/minimax.md | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/docs/providers/minimax.md b/docs/providers/minimax.md index f9393e50e53..fd3309c2d9a 100644 --- a/docs/providers/minimax.md +++ b/docs/providers/minimax.md @@ -256,6 +256,12 @@ The bundled `minimax` plugin registers MiniMax T2A v2 as a speech provider for MP3 to 48kHz Opus with `ffmpeg`, because the Feishu/Lark file API only accepts `file_type: "opus"` for native audio messages. +| Setting | Env var | Default | Description | +| ---------------------------------------- | ---------------------- | ----------------------------- | -------------------------------- | +| `messages.tts.providers.minimax.baseUrl` | `MINIMAX_API_HOST` | `https://api.minimax.io` | MiniMax T2A API host. | +| `messages.tts.providers.minimax.model` | `MINIMAX_TTS_MODEL` | `speech-2.8-hd` | TTS model id. | +| `messages.tts.providers.minimax.voiceId` | `MINIMAX_TTS_VOICE_ID` | `English_expressive_narrator` | Voice id used for speech output. | + ### Music generation The bundled `minimax` plugin also registers music generation through the shared