For the last few years, I couldn’t visit the conferences due to Covid-19, but I’ll present at the on-site conference this time for INTERSPEECH 2022 in Incheon, South Korea.
I also participate in this conference as a session chair and present 2 papers in oral sessions.

The first paper is “Learning Audio-Text Agreement for Open-vocabulary Keyword Spotting”, where I collaborated with DSP&AI Lab., at Yonsei University. This paper proposes an end-to-end open-vocabulary keyword spotting using cross-attention mechanism between audio and text modalities.

The second paper is “SASV 2022: The First Spoofing-Aware Speaker Verification Challenge.” It is for the SASV challenge at INTERSPEECH 2022, and the paper demonstrates and compares the proposed methods of participants.

If you are interested in these works, please visit the link below.

[Learning Audio-Text Agreement for Open-vocabulary Keyword Spotting]
[SASV 2022: The First Spoofing-Aware Speaker Verification Challenge]