mirror of
https://github.com/xinnan-tech/xiaozhi-esp32-server.git
synced 2026-07-26 09:03:54 +08:00
fix: 停止帧发送后与原先音频时序问题
回退退出工具打断处理(在大模型思考期间进行打断,然后大模型立马调用退出工具会造成死锁,因模型思考时间和是否调用工具不确定性,状态难管理)
This commit is contained in:
@@ -15,8 +15,6 @@ TAG = __name__
|
||||
|
||||
|
||||
async def handleAudioMessage(conn: "ConnectionHandler", audio):
|
||||
if conn.is_exiting:
|
||||
return
|
||||
# 当前片段是否有人说话
|
||||
have_voice = conn.vad.is_vad(conn, audio)
|
||||
# 如果设备刚刚被唤醒,短暂忽略VAD检测
|
||||
|
||||
Reference in New Issue
Block a user