With her
Power on, speak or type. One emotion tag drives her voice, her face and her motion — in sync.
No Live2D model yet
Her model comes with her voice: run
(README, “Set her up”), then reload this page.
Any other Cubism model works too: put its folder in
envs\brain\Scripts\python.exe scripts\setup_tts.py(README, “Set her up”), then reload this page.
Any other Cubism model works too: put its folder in
avatar/models/.
Blur
12px
Your camera
Tool dialogue what she asks her helpers, and what they say back
When she hands something to Hermes or nanoMuse, her request and everything that comes back shows up here.
Session
Session cost
Amika
—
her own calls · display period
Agents
—
Hermes, nanoMuse · display period
First voice
—
avg, last 10 rounds
Prompt cache hit
—
prompt tokens read from cache
Latency and cache per round
Rounds appear here as you talk.
Her helpers jobs in flight · anything waiting for your yes
Transcript会話
Nothing yet — say hi.