International Conference on Acoustics, Speech, and Signal Processing

Well, we first, uh, studied this kind of like vulnerability of the automatic speaker recognition systems.

And then circa 2012, if I still remember the year correct, in ICASSP, I met with Nick Evans, who had been independently working on similar topics in more general biometric face recognition spoofing and others.

And I also got connected with Junichi Yamagishi, and basically the three of us then organized in Interspeech 2013.

Uh, it wasn't really a challenge, but more like a special session for us to, to see what the reaction of the community is.

There's work on semantic voice activity detection as a dialogue manager, predicting control tokens to regulate turn-switching.

The HumDial Challenge at ICASSP this year released a dual-channel dataset of real conversations with interruptions and overlapping speech.

All of it's about systems that listen and speak at the same time.

And what does the ECHO result actually say? It says most full-duplex systems have a pronounced bias toward yield.