Andreas Nautsch

Andreas Nautsch, M.Sc.

Position: Ph.D. Student (former member)

Andreas Nautsch started studying computer science in 2008 at Hochschule Darmstadt (h_da) in a cooperative study program, he received his B.Sc. and M.Sc. degrees in 2012 and 2014, respectively. From 2008 to 2014 he worked at atip GmbH, Frankfurt am Main, as Software Engineer with focus on automatic speech recognition and speaker recognition for interactive voice response systems. In 2013, he went to an exchange semester at Mälardalens Högskola (MDH), Västerås, Sweden. In June 2014 he became Ph.D. student member of da/sec at the Center for Advanced Security Research Darmstadt (CASED), which became the Center for Research in Security and Privacy (CRISP). Since April 2015, Andreas is formal Ph.D. student of Prof. Dr. Max Mühlhäuser at the Technische Universität Darmstadt (TU Darmstadt) with research supervision of Prof. Dr. Christoph Busch at the Hochschule Darmstadt (h_da). He went to research stays at the ATVS biometrics group at Universidad Autónoma de Madrid, Spain from December 2015 to March 2016, to Voxalys AB in Gothenburg, Sweden, during March 2017, and to the Computer Vision lab at the University of Nottingham from September to December 2017, respectively.

Research Interests

  • Speaker Recognition
  • Information Theory
  • Bayesian Statistics
  • Machine Learning
  • Biometrics

Activities/Memberships

Publications

Journal
Conferences
  • Andreas Nautsch, Sergey Isadskiy, Jascha Kolberg, Marta Gomez-Barrero, Christoph Busch, „Homomorphic Encryption for Speaker Recognition: Protection of Biometric Templates and Vendor Model Parameters“, in Proceedings Speaker Odyssey, 2018. [PDF]
  • Andreas Nautsch, Didier Meuwly, Daniel Ramos, Jonas Lindh, Christoph Busch, „Making Likelihood Ratios Digestible for Cross-Application Performance Assessment“, in International Conference on Acoustics, Speech and Signal Processing (ICASSP’18), 2018. [PDF] [code] [poster]
  • Andreas Nautsch, Søren Trads Steen, Christoph Busch, „Deep Quality-informed Score Normalization for Privacy-friendly Speaker Recognition in unconstrained Environments“, In Proceedings of International Conference of the Biometrics Special Interest Group (BIOSIG’17), 2017. [PDF]
  • Ulrich Scherhag, Andreas Nautsch, Christian Rathgeb, Marta Gomez-Barrero, Raymond N. J. Veldhuis, Luuk Spreeuwers, Maikel Schils, Davide Maltoni, Patrick Grother, Sebastien Marcel, Ralph Breithaupt, Raghavendra Ramachandra, Christoph Busch, „Biometric Systems under Morphing Attacks: Assessment of Morphing Techniques and Vulnerability Reporting“, in Proceedings of BIOSIG (BIOSIG’17) 2017. [PDF]
  • Kong Aik Lee, Ville Hautamäki, Tomi Kinnunen, Anthony Larcher, Chunlei Zhang, Andreas Nautsch, Themos Stafylakis et al., „The I4U Mega Fusion and Collaboration for NIST Speaker Recognition Evaluation 2016“, In Proceedings of INTERSPEECH, 2017. [PDF]
  • Jonas Lindh, Andreas Nautsch, Therese Leinonen, and Joel Åkesson, „Comparison Between Perceptual and Automatic Systems on Finnish Phone Speech Data (FinEval1) – a pilot test using score simulations“, In Proceedings of IAFPA, 2017.
  • Andreas Nautsch, Reiner Bamberger, Christoph Busch, „Decision Robustness of Voice Activity Segmentation in unconstrained mobile Speaker Recognition Environments“, In Proceedings of International Conference of the Biometrics Special Interest Group (BIOSIG’16), 2016. [PDF]
  • Ulrich Scherhag, Andreas Nautsch, Christian Rathgeb, Christoph Busch, „Unit-Selection Attack Detection Based on Unfiltered Frequency-Domain Features“, In Proceedings of INTERSPEECH (IS’16), 2016. [PDF]
  • Andreas Nautsch, Rahim Saeidi, Christian Rathgeb, Christoph Busch, „Robustness of Quality-based Score Calibration of Speaker Recognition Systems with respect to low-SNR and short-duration conditions“, In Proceedings of Odyssey 2016: The Speaker and Language Recognition Workshop (Odyssey’16), 2016. [PDF]
  • Marco Paulini, Christian Rathgeb, Andreas Nautsch, Hermine Reichau, Herbert Reininger, Christoph Busch, „Multi-Bit Allocation: Preparing Voice Biometrics for Template Protection“, In Proceedings of Odyssey 2016: The Speaker and Language Recognition Workshop (Odyssey’16), 2016. [PDF]
  • Andreas Nautsch, Hong Hao, Themos Stafylakis, Christian Rathgeb, Christoph Busch, „Towards PLDA-RBM based Speaker Recognition in Mobile Environment: Designing Stacked/Deep PLDA-RBM Systems“, In Proceedings of International Conference on Acoustics, Speech and Signal Processing (ICASSP’16), 2016. [PDF] [slides]
  • Andreas Nautsch, Rahim Saeidi, Christian Rathgeb, Christoph Busch, „Analysis of mutual duration and noise effects in speaker recognition: benefits of condition-matched cohort selection in score normalization“, In Proceedings of INTERSPEECH (IS’15), 2015. [PDF] [poster]
  • Andreas Nautsch, Christian Rathgeb, Rahim Saeidi, Christoph Busch, „Entropy analysis of i-vector feature spaces in duration-sensitive speaker recognition“, In Proceedings of International Conference on Acoustics, Speech and Signal Processing (ICASSP’15), 2015. [PDF] [poster]
Technical Reports
  • Andreas Nautsch, Ulrich Scherhag, Sergey Isadskiy, Christian Rathgeb, Christoph Busch, „da/sec ASVspoof2017 Submission“, In Proceedings of ASVspoof 2017 evaluation, 2017.
  • Andreas Nautsch, Sergey Isadskiy, Christian Rathgeb, Christoph Busch, „HDA NIST SRE’16 Systems (CRISP & I4U)“, In Proceedings of NIST SRE’16 workshop, 2016.
  • Kong Aik Lee et al., „The I4U submission to the 2016 NIST Speaker Recognition Evaluation“, In Proceedings of NIST SRE’16 workshop, 2016.
  • Andreas Nautsch, Alicia Lozano Diez, Daniel Ramos Castro, „Full-Posterior PLDA in Speaker Recognition: Technical Literature Review“, Technical Report, (released in 2017), 2016. [PDF]
Pre-PhD studies
  • Andreas Nautsch, Christian Rathgeb, Christoph Busch, „Bridging Gaps: An Application of Feature Warping to Online Signature Verification“, In Proceedings of International Carnahan Conference in Security Technology (ICCST’14), 2014. [slides]
  • Andreas Nautsch, Christian Rathgeb, Christoph Busch, Herbert Reininger, Klaus Kasper, „Towards Duration Invariance of i-Vector-based Adaptive Score Normalization“, In Proceedings of Odyssey: The Speaker and Language Recognition Workshop (ODYSSEY’14), pp. 60-67, 2014. [PDF] [poster]
  • Andreas Nautsch, Klaus Kasper, Herbert Reininger, Martin Wagner, Lubos Kresja, „atip GmbH NIST SRE12 Submission“, In Proceedings of NIST SRE’12 workshop, 2012.
  • Andreas Nautsch, Anne Schönwandt, Klaus Kasper, Herbert Reininger, Martin Wagner, „Improving channel robustness in text-independent speaker verification using adaptive virtual cohort models“, In Proceedings of the International Conference of the Biometrics Special Interest Group (BIOSIG’12), pp. 1-5, 2012. [PDF]
  • Andreas Nautsch, „Ohr-Erkennung – Stand der Technik“, In Proceedings of the International Conference of the Biometrics Special Interest Group (BIOSIG’11), pp. 275-282, 2011. [PDF]

Distinctions

Theses

  • Master of Science, „Speaker Verification using i-Vectors“, Hochschule Darmstadt, 2014. [PDF]
  • Bachelor of Science, „Score-Normierung in der Sprecherverifikation mittels Kohortensprechern: Evaluation verschiedener Ansätze zur Synthese virtueller Kohortensprecher im Hinblick auf die biometrische Performanz“, Hochschule Darmstadt, AV Akademiker Verlag, ISBN: 9783639459548, 2012.

Further Talks

  • 2018-05-15, „Technischer Datenschutz in der Sprecher-Erkennung“, 1. Campus Day: „Voice / Voicification“, Telekom, Darmstadt, Germany.
  • 2017-11-16, „Speaker Recognition in unconstrained Environments“, reading group seminar of the computer vision lab at the University of Nottingham, Nottingham, UK.
  • 2017-10-27, „Speaker Recognition in unconstrained Environments“, Voice, Lips and Face: A concept workshop, Oxford, UK.
  • 2017-08-23, „Selection challenges: datasets & hypotheses“, Special Session: „Speaker Comparison for Forensic and Investigative Applications III“ (panel and discussion), Interspeech, Stockholm, Sweden.
  • 2017-06-27, „Technical Standardization Activities related to Voice Biometrics“, project meeting OCTAVE: Objective Control for TAlker VErification, Florence, Italy.
  • 2017-04-21, „Speaker Recognition in unconstrained Environments“, project meeting BATL: Biometric Authentication with a Timeless Learner, Darmstadt, Germany.
  • 2017-03-23, „Speaker Recognition in unconstrained Environments“, machine learning seminar at the Chalmers University of Technology, Gothenburg, Sweden.
  • 2017-03-21, „Speaker Recognition in unconstrained Environments“, linguistics seminar at the University of Gothenburg, Gothenburg, Sweden.
  • 2016-06-18, „Methodology and validation tools for evaluation of performance of computer based methods used in forensic biometrics“, NWIP 29795-8, ISO/IEC JTC 1/SC 37 WG 5, Paris, France. Presentation by Dr. Rudolf Haraksim, in cooperation with Prof. Didier Meuwly.
  • 2016-05-03, „Making decisions with biometric systems: the usefulness of a Bayesian perspective“, NIST International Biometric Performance Testing Conference 2016 (IBPC’16), Gaithersburg, MD, USA. Shared presentation with Prof. Daniel Ramos. [slides]
  • 2016-02-01, „Speaker Recognition in unconstrained Environments“, seminar of the ATVS Biometrics Research Group at UAM, Madrid, Spain.
  • 2015-12-15, „Entropy Analysis in Speaker Recognition“, German TeleTrust Biometrics Working Group, Berlin, Germany. [slides]
  • 2015-06-22, „Future biometric performance testing“, Proposed contents for revising ISO/IEC IS 19795-1, ISO/IEC JTC 1/SC 37 joint WG 1, 2, 3 & 5 meeting, Gjøvik, Norway.
  • 2015-06-20, „19795-1: Need for amendment“, Proposed contents for revising ISO/IEC IS 19795-1, ISO/IEC JTC 1/SC 37 WG 5, Gjøvik, Norway.

Science & education @ YouTube (endorsed short list)

Disclaimer: opinions expressed here or implied by links provided are personal,
i.e. they do not represent the official views of the Hochschule Darmstadt.

misc. appetizers