GeminiVRM v0.2.0 Podcast Mode Guide

GeminiVRM v0.2.0 release header

What Changes For You

One topic can now drive a capped Yukito/Kiyoka back-and-forth instead of only a single-avatar reply flow.
Podcast mode has its own turn cap and voice routing, so you can keep single-character chat settings separate from dual-host experiments.
The viewer runtime feels more alive because bundled Mixamo idle clips now rotate by default and talking motion swaps in during speech.

Open Settings and switch Conversation mode to Podcast mode.
Wait for both bundled hosts to finish loading.
Open Podcast settings if you want to change the loop cap or assign different prebuilt voices to Yukito and Kiyoka.
Enter one topic and send it.
Watch the stage alternate speakers until the configured turn cap is reached.

Podcast mode is intentionally turn-based. It does not try to overlap both hosts speaking at once.
Later turns reuse the previous speaker's audio as Gemini Live input when possible.
If audio relay reuse fails, GeminiVRM falls back to transcript-driven continuation instead of cancelling the run.
Motion changes apply across chat and podcast mode, with Random Idle now rotating the bundled Mixamo clips.

Browser-side automation can query state or drive sends through window.geminiVrmControl.
The postMessage bridge can switch modes, update podcast settings, change motion presets, and reset the active log.
Production use still requires the external-control gate to be explicitly enabled.

npm run lint
npm run build
npm run docs:build
$env:BASE_PATH='/GeminiVRM'; npm run build:pages
npm run e2e:smoke
powershell -ExecutionPolicy Bypass -File D:\Prj\gh-release-notes-skill\scripts\verify-svg-assets.ps1 -RepoPath . -Path public/favicon.svg,docs/public/releases/release-header-v0.2.0.svg
npm run verify:release-header-layout -- --input docs/public/releases/release-header-v0.2.0.svg
powershell -ExecutionPolicy Bypass -File D:\Prj\gh-release-notes-skill\scripts\verify-release-qa-inventory.ps1 -RepoPath . -Tag v0.2.0