no code implementations • 31 May 2023 • Hye-jin Shim, Rosa González Hautamäki, Md Sahidullah, Tomi Kinnunen
Shortcut learning, or `Clever Hans effect` refers to situations where a learning agent (e. g., deep neural networks) learns spurious correlations present in data, resulting in biased models.
1 code implementation • 31 Mar 2022 • Lauri Tavi, Tomi Kinnunen, Rosa González Hautamäki
Due to a constantly increasing amount of speech data that is stored in different types of databases, voice privacy has become a major concern.
no code implementations • 26 Mar 2021 • Bhusan Chettri, Rosa González Hautamäki, Md Sahidullah, Tomi Kinnunen
Voice anti-spoofing aims at classifying a given utterance either as a bonafide human sample, or a spoofing attack (e. g. synthetic or replayed sample).
no code implementations • 3 Jun 2019 • Ville Vestman, Tomi Kinnunen, Rosa González Hautamäki, Md Sahidullah
Our goal is to gain insights how well similarity rankings transfer from the attacker's ASV system to the attacked ASV system, whether the attackers are able to improve their attacks by mimicking, and how the properties of the voices of attackers change due to mimicking.
no code implementations • 9 Nov 2018 • Tomi Kinnunen, Rosa González Hautamäki, Ville Vestman, Md Sahidullah
We consider technology-assisted mimicry attacks in the context of automatic speaker verification (ASV).