multimodal inference · text / image / audio → text + audio
text_ttft
—
audio_ttft
—
status
idle
default
default
Call
from file
take photo
cancel
capture
Clone Voice
请自然朗读下面这句话。建议 3 到 6 秒，环境尽量安静。录完后会自动做 1.5 倍速、提取 `CAM++` speaker embedding 与 `Mimi` ref codes，然后立即加入音色列表。
今天阳光很好，欢迎来到我的语音克隆小实验。
点击 start 开始录音
cancel
start
live voice · bidirectional streaming
default
log
connecting
init
00:00:00
latest · awaiting first turn
speak naturally — model responds with voice and text
cam
end
session transcript
0 turns
× close
