Tandem Assessment of Spoofing Countermeasures and Automatic Speaker Verification: Fundamentals

• URL: http://arxiv.org/abs/2007.05979v2
• Date: Tue, 25 Aug 2020 10:40:50 GMT
• Title: Tandem Assessment of Spoofing Countermeasures and Automatic Speaker Verification: Fundamentals
• Authors: Tomi Kinnunen and H\'ector Delgado and Nicholas Evans and Kong Aik Lee and Ville Vestman and Andreas Nautsch and Massimiliano Todisco and Xin Wang and Md Sahidullah and Junichi Yamagishi and Douglas A. Reynolds
• Abstract summary: The reliability of spoofing countermeasures (CMs) is gauged using the equal error rate (EER) metric. This paper presents several new extensions to the tandem detection cost function (t-DCF) It is hoped that adoption of the t-DCF for the CM assessment will help to foster closer collaboration between the anti-spoofing and ASV research communities.
• Score: 59.34844017757795
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent years have seen growing efforts to develop spoofing countermeasures (CMs) to protect automatic speaker verification (ASV) systems from being deceived by manipulated or artificial inputs. The reliability of spoofing CMs is typically gauged using the equal error rate (EER) metric. The primitive EER fails to reflect application requirements and the impact of spoofing and CMs upon ASV and its use as a primary metric in traditional ASV research has long been abandoned in favour of risk-based approaches to assessment. This paper presents several new extensions to the tandem detection cost function (t-DCF), a recent risk-based approach to assess the reliability of spoofing CMs deployed in tandem with an ASV system. Extensions include a simplified version of the t-DCF with fewer parameters, an analysis of a special case for a fixed ASV system, simulations which give original insights into its interpretation and new analyses using the ASVspoof 2019 database. It is hoped that adoption of the t-DCF for the CM assessment will help to foster closer collaboration between the anti-spoofing and ASV research communities.

Related papers

• ASVspoof 5: Crowdsourced Speech Data, Deepfakes, and Adversarial Attacks at Scale [59.25180900687571]
  ASVspoof 5 is the fifth edition in a series of challenges that promote the study of speech spoofing and deepfake attacks. We describe the two challenge tracks, the new database, the evaluation metrics, and the evaluation platform, and present a summary of the results.
  arXiv  Detail & Related papers  (2024-08-16T13:37:20Z)
• To what extent can ASV systems naturally defend against spoofing attacks? [73.0766904568922]
  This study investigates whether ASV effortlessly acquires robustness against spoofing attacks. We demonstrate that the evolution of ASV inherently incorporates defense mechanisms against spoofing attacks.
  arXiv  Detail & Related papers  (2024-06-08T03:44:39Z)
• a-DCF: an architecture agnostic metric with application to spoofing-robust speaker verification [21.428968328957897]
  We propose an architecture-agnostic detection cost function (a-DCF) A-DCF reflects the cost of decisions in a Bayes risk sense, with explicitly defined class priors and detection cost model. We demonstrate the merit of the a-DCF through the benchmarking evaluation of architecturally-heterogeneous spoofing-robust ASV solutions.
  arXiv  Detail & Related papers  (2024-03-03T00:58:27Z)
• Generalizing Speaker Verification for Spoof Awareness in the Embedding Space [30.094557217931563]
  ASV systems can be spoofed using various types of adversaries. We propose a novel yet simple backend classifier based on deep neural networks. Experiments are conducted on the ASVspoof 2019 logical access dataset.
  arXiv  Detail & Related papers  (2024-01-20T07:30:22Z)
• Voice Spoofing Countermeasures: Taxonomy, State-of-the-art, experimental analysis of generalizability, open challenges, and the way forward [2.393661358372807]
  We conduct a review of the literature on spoofing detection using hand-crafted features, deep learning, end-to-end, and universal spoofing countermeasure solutions. We report the performance of these countermeasures on several datasets and evaluate them across corpora.
  arXiv  Detail & Related papers  (2022-10-02T03:53:37Z)
• Tackling Spoofing-Aware Speaker Verification with Multi-Model Fusion [88.34134732217416]
  This work focuses on fusion-based SASV solutions and proposes a multi-model fusion framework to leverage the power of multiple state-of-the-art ASV and CM models. The proposed framework vastly improves the SASV-EER from 8.75% to 1.17%, which is 86% relative improvement compared to the best baseline system in the SASV challenge.
  arXiv  Detail & Related papers  (2022-06-18T06:41:06Z)
• Spoofing-Aware Speaker Verification by Multi-Level Fusion [86.19341932163813]
  A spoofing aware speaker verification (SASV) challenge aims to facilitate the research of integrated CM and ASV models. We propose a novel multi-model and multi-level fusion strategy to tackle the SASV task.
  arXiv  Detail & Related papers  (2022-03-29T09:16:38Z)
• Optimizing Tandem Speaker Verification and Anti-Spoofing Systems [45.66319648049384]
  We propose to optimize the tandem system directly by creating a differentiable version of t-DCF and employing techniques from reinforcement learning. Results indicate that these approaches offer better outcomes than finetuning, with our method providing a 20% relative improvement in the t-DCF in the ASVSpoof19 dataset.
  arXiv  Detail & Related papers  (2022-01-24T14:27:28Z)
• Investigating Robustness of Adversarial Samples Detection for Automatic Speaker Verification [78.51092318750102]
  This work proposes to defend ASV systems against adversarial attacks with a separate detection network. A VGG-like binary classification detector is introduced and demonstrated to be effective on detecting adversarial samples.
  arXiv  Detail & Related papers  (2020-06-11T04:31:56Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.