(minimax): add new TTS models

[tinalenguyen]

No description provided.

[devin-ai-integration]

✅ Devin Review: No Issues Found

Devin Review analyzed this PR and found no potential bugs to report.

View in Devin Review to see 1 additional finding.

[toubatbrian]

This is an automated Claude Code Routine created by @toubatbrian. Right now it is in an experimentation stage. The automation will start porting this PR into agents-js automatically.

---

Porting status: SKIPPED

After inspecting livekit/agents-js, there is no minimax plugin present in the plugins/ tree, so there is no existing TTS module to extend with the new speech-2.8-hd / speech-2.8-turbo model literals. Adding only the model strings without a corresponding @livekit/agents-plugin-minimax package would be a no-op in JS. Creating a brand-new Minimax plugin from scratch is out of scope for this small model-name update.

No JS PR will be opened for this change. If/when a Minimax plugin is added to agents-js, these model names should be included at that time.

cc @toubatbrian @livekit/agent-devs

---

Generated by Claude Code

[toubatbrian]

Update: per follow-up instruction, a full MiniMax plugin has now been created in agents-js to serve as the target for this PR's model additions (and future MiniMax work).

JS PR: livekit/agents-js#1287

The new plugin ships with speech-2.8-hd and speech-2.8-turbo included in the TTSModel type literal, along with the rest of the MiniMax TTS API surface ported from livekit-plugins-minimax. See the JS PR description for the full list of implementation differences (most notably: PCM-only output in the initial port, text_pacing omitted).

cc @toubatbrian @livekit/agent-devs

---

Generated by Claude Code