JAIST Repository >
School of Information Science >
Conference Papers >
Conference Papers >

Please use this identifier to cite or link to this item: http://hdl.handle.net/10119/18158

Title: Data Augmentation Using McAdams-Coefficient-Based Speaker Anonymization for Fake Audio Detection
Authors: Li, Kai
Li, Sheng
Lu, Xugang
Akagi, Masato
Liu, Meng
Zhang, Lin
Zeng, Chang
Wang, Longbiao
Dang, Jianwu
Unoki, Masashi
Keywords: fake audio detection
data augmentation
McAdams coefficients
speaker anonymization
Issue Date: 2022-09
Publisher: International Speech Communication Association
Magazine name: Proc. InterSpeech 2022
Start page: 664
End page: 668
DOI: 10.21437/Interspeech.2022-10088
Abstract: Fake audio detection (FAD) is a technique to distinguish synthetic speech from natural speech. In most FAD systems, removing irrelevant features from acoustic speech while keeping only robust discriminative features is essential. Intuitively, speaker information entangled in acoustic speech should be suppressed for the FAD task. Particularly in a deep neural network (DNN)-based FAD system, the learning system may learn speaker information from a training dataset and cannot generalize well on a testing dataset. In this paper, we propose to use the speaker anonymization (SA) technique to suppress speaker information from acoustic speech before inputting it into a DNN-based FAD system. We adopted the McAdamscoefficient-based SA (MC-SA) algorithm, and this is expected that the entangled speaker information will not be involved in the DNN-based FAD learning. Based on this idea, we implemented a light convolutional neural network bidirectional long short-term memory (LCNN-BLSTM)-based FAD system and conducted experiments on the Audio Deep Synthesis Detection Challenge (ADD2022) datasets. The results showed that removing the speaker information from acoustic speech improved the relative performance in the first track of ADD2022 by 17.66%.
Rights: Copyright (C) 2022 International Speech Communication Association. Kai Li, Sheng Li, Xugang Lu, Masato Akagi, Meng Liu, Lin Zhang, Chang Zeng, Longbiao Wang, Jianwu Dang, Masashi Unoki, Proc. InterSpeech2022, 2022, pp.664-668. doi:10.21437/Interspeech.2022-10088
URI: http://hdl.handle.net/10119/18158
Material Type: publisher
Appears in Collections:b11-1. 会議発表論文・発表資料 (Conference Papers)

Files in This Item:

File Description SizeFormat
li22o_interspeech.pdf295KbAdobe PDFView/Open

All items in DSpace are protected by copyright, with all rights reserved.

 


Contact : Library Information Section, Japan Advanced Institute of Science and Technology