POCO: A Voice Spoofing and Liveness Detection Corpus Based on Pop Noise

Kosuke Akimoto, Seng Pei Liew, Sakiko Mishima, Ryo Mizushima, Kong Aik Lee

Abstract: We present a new database of voice recordings with the goal of promoting research on protection of automatic speaker verification systems from voice spoofing, such as replay attacks. Specifically, we focus on the liveness feature of live speech, i.e., pop noise, and the corresponding voice recordings without this feature, for the purpose of combating spoofing via liveness detection. Our database includes simultaneous recordings using a microphone array, as well as recordings at various distances and positions. To the best of our knowledge, this is the first publicly available database that has been particularly designed to study the liveness features of voice recordings under various conditions.1