ASVspoof 2017 CHALLENGE: Audio Replay Detection for Automatic Speaker Verification Anti-spoofing (Md Sahidullah )


Subject: ASVspoof 2017 CHALLENGE: Audio Replay Detection for Automatic Speaker Verification Anti-spoofing
From:    Md Sahidullah  <sahidullahmd@xxxxxxxx>
Date:    Fri, 23 Dec 2016 21:29:01 +0200
List-Archive:<http://lists.mcgill.ca/scripts/wa.exe?LIST=AUDITORY>

--001a11372e92f55b540544586593 Content-Type: text/plain; charset=UTF-8 [Apologies for cross-postings] =*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=* ASVspoof 2017 CHALLENGE: Audio replay detection for automatic speaker verification anti-spoofing http://www.spoofingchallenge.org/ =*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=* Are you good at machine learning for audio signals? Are you good at discriminating 'fake' signals from authentic ones? Are you looking for new audio processing challenges? Do you work in the domain of speaker recognition? ASVspoof 2017 challenge might be for you! CHALLENGE TASK: Given a short clip of speech audio, determine whether it contains a GENUINE human voice (live recording), or a REPLAY recording (fake). You will be provided a development set containing genuine/replay labeled audio examples, along with further metadata such as speech content and the devices used in the replay recordings. Your task is to develop a system that assigns a single 'liveness' or 'genuineness' score value to new audio samples and to execute that system on a set of test files for which the ground truth is not provided. We provide a Matlab-based reference baseline method to kick-off quickly towards developing your new ideas! For more details, refer to the evaluation plan on the website: http://www.spoofingchallenge.org/ BACKGROUND: The goal of the challenge series is to enhance the security of automatic speaker verification (ASV) systems from being intentionally circumvented using fake recordings, also known as 'spoofing attacks' or 'representation attacks' in the context of biometrics. ASVspoof 2017 is the second edition of a challenge kicked off in 2015, and the new perspective in ASVspoof 2017 are the replay attacks, especially 'unseen' attacks - for instance, containing replay environments, devices, and speakers that might be very different from those in the development data. Despite 'ASV' being in the challenge title, you do NOT require knowledge of automatic speaker verification: the task is a 'standalone' replay audio detection task that can be addressed as a generic acoustic pattern classification problem. We welcome as many new ideas to the problem as possible! SCHEDULE: Development data published: December 23rd, 2016 Evaluation data published: February 10, 2017 Evaluation set scores due: February 24, 2017 Results available: March 3, 2017 Interspeech paper deadline: March 14, 2017 Metadata/keys published: May 2017 Interspeech special session: August 2017 REGISTRATION: Send a free-worded e-mail to asvspoof2017@xxxxxxxx to register and obtain the dev data. ORGANIZERS: Tomi Kinnunen, University of Eastern Finland, FINLAND Nicholas Evans, Eurecom, FRANCE Junichi Yamagishi, University of Edinburgh, UK Kong Aik Lee, Institute for Infocomm Research, SINGAPORE Md Sahidullah, University of Eastern Finland, FINLAND Massimiliano Todisco, Eurecom, FRANCE Hector Delgado, Eurecom, FRANCE CONTACT: asvspoof2017@xxxxxxxx -- Dr. Md Sahidullah Post-doctoral researcher School of Computing University of Eastern Finland Joensuu, P.O. Box 111 Finland- 80101 tel. +358-466250731 website: http://cs.joensuu.fi/~sahid/ --001a11372e92f55b540544586593 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable <div dir=3D"ltr"><div style=3D"font-size:12.8px"><span style=3D"font-size:1= 2.8px">[Apologies for cross-postings]</span><br></div><div style=3D"font-si= ze:12.8px"><span style=3D"font-size:12.8px"><br></span></div><div style=3D"= font-size:12.8px">=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*= =3D*<wbr>=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*<wbr>= =3D*=3D*=3D*=3D*</div><div style=3D"font-size:12.8px"><br></div><div style= =3D"font-size:12.8px">=C2=A0 ASVspoof 2017 CHALLENGE:</div><div style=3D"fo= nt-size:12.8px">=C2=A0 Audio replay detection for automatic speaker verific= ation anti-spoofing</div><div style=3D"font-size:12.8px"><br></div><div sty= le=3D"font-size:12.8px">=C2=A0=C2=A0<a href=3D"http://www.spoofingchallenge= .org/" target=3D"_blank">http://www.spoofingchallenge.<wbr>org/</a></div><d= iv style=3D"font-size:12.8px"><br></div><div style=3D"font-size:12.8px">=3D= *=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*<wbr>=3D*=3D*=3D*= =3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*<wbr>=3D*=3D*=3D*=3D*</div>= <div style=3D"font-size:12.8px"><br></div><div style=3D"font-size:12.8px">= =C2=A0 Are you good at machine learning for audio signals? Are you good at = discriminating &#39;fake&#39; signals from authentic ones? Are you looking = for new audio processing challenges? Do you work in the domain of speaker r= ecognition?</div><div style=3D"font-size:12.8px">=C2=A0 ASVspoof 2017 chall= enge might be for you!</div><div style=3D"font-size:12.8px"><br></div><div = style=3D"font-size:12.8px">CHALLENGE TASK:</div><div style=3D"font-size:12.= 8px"><br></div><div style=3D"font-size:12.8px">=C2=A0 Given a short clip of= speech audio, determine whether it contains</div><div style=3D"font-size:1= 2.8px">=C2=A0 a GENUINE human voice (live recording), or a REPLAY recording= (fake).</div><div style=3D"font-size:12.8px"><br></div><div style=3D"font-= size:12.8px">=C2=A0 You will be provided a development set containing genui= ne/replay labeled audio examples, along with further metadata such as speec= h content and the devices used in the replay recordings. Your task is to de= velop a system that assigns a</div><div style=3D"font-size:12.8px">=C2=A0 s= ingle &#39;liveness&#39; or &#39;genuineness&#39; score value to new audio = samples and to execute that system on a set of test files for which the gro= und truth is not provided. We provide a Matlab-based reference baseline met= hod to kick-off quickly towards developing your new ideas!</div><div style= =3D"font-size:12.8px"><br></div><div style=3D"font-size:12.8px">=C2=A0 For = more details, refer to the evaluation plan on the website:</div><div style= =3D"font-size:12.8px">=C2=A0=C2=A0<a href=3D"http://www.spoofingchallenge.o= rg/" target=3D"_blank">http://www.spoofingchallenge.<wbr>org/</a></div><div= style=3D"font-size:12.8px"><br></div><div style=3D"font-size:12.8px">BACKG= ROUND:</div><div style=3D"font-size:12.8px"><br></div><div style=3D"font-si= ze:12.8px">=C2=A0 The goal of the challenge series is to enhance the securi= ty of automatic speaker verification (ASV) systems from being intentionally= circumvented using fake recordings, also known as &#39;spoofing attacks&#3= 9; or &#39;representation attacks&#39; in the context of biometrics. ASVspo= of 2017 is the second edition of a challenge kicked off in 2015, and the ne= w perspective in ASVspoof 2017 are the replay attacks, especially &#39;unse= en&#39; attacks - for instance, containing replay environments, devices, an= d speakers that might be very different from those in the development data.= </div><div style=3D"font-size:12.8px"><br></div><div style=3D"font-size:12.= 8px">=C2=A0 Despite &#39;ASV&#39; being in the challenge title, you do NOT = require knowledge of automatic speaker verification: the task is a &#39;sta= ndalone&#39; replay audio detection task that can be addressed as a generic= acoustic pattern classification problem. We welcome as many new ideas to t= he problem as possible!</div><div style=3D"font-size:12.8px"><br></div><div= style=3D"font-size:12.8px">SCHEDULE:</div><div style=3D"font-size:12.8px">= <br></div><div style=3D"font-size:12.8px">=C2=A0 Development data published= : =C2=A0 December 23rd, 2016</div><div style=3D"font-size:12.8px">=C2=A0 Ev= aluation data published: =C2=A0 =C2=A0February 10, 2017</div><div style=3D"= font-size:12.8px">=C2=A0 Evaluation set scores due: =C2=A0 =C2=A0February 2= 4, 2017</div><div style=3D"font-size:12.8px">=C2=A0 Results available: =C2= =A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0March 3, 2017</div><div style=3D"font= -size:12.8px">=C2=A0 Interspeech paper deadline: =C2=A0 March 14, 2017</div= ><div style=3D"font-size:12.8px">=C2=A0 Metadata/keys published: =C2=A0 =C2= =A0 =C2=A0May 2017</div><div style=3D"font-size:12.8px">=C2=A0 Interspeech = special session: =C2=A0August 2017</div><div style=3D"font-size:12.8px"><br= ></div><div style=3D"font-size:12.8px">REGISTRATION:</div><div style=3D"fon= t-size:12.8px"><br></div><div style=3D"font-size:12.8px">=C2=A0 Send a free= -worded e-mail to=C2=A0<a href=3D"mailto:asvspoof2017@xxxxxxxx" target=3D"= _blank">asvspoof2017@xxxxxxxx</a></div><div style=3D"font-size:12.8px">=C2= =A0 to register and obtain the dev data.</div><div style=3D"font-size:12.8p= x"><br></div><div style=3D"font-size:12.8px">ORGANIZERS:</div><div style=3D= "font-size:12.8px"><br></div><div style=3D"font-size:12.8px">=C2=A0 Tomi Ki= nnunen, University of Eastern Finland, FINLAND</div><div style=3D"font-size= :12.8px">=C2=A0 Nicholas Evans, Eurecom, FRANCE</div><div style=3D"font-siz= e:12.8px">=C2=A0 Junichi Yamagishi, University of Edinburgh, UK</div><div s= tyle=3D"font-size:12.8px">=C2=A0 Kong Aik Lee, Institute for Infocomm Resea= rch, SINGAPORE</div><div style=3D"font-size:12.8px">=C2=A0 Md Sahidullah, U= niversity of Eastern Finland, FINLAND</div><div style=3D"font-size:12.8px">= =C2=A0 Massimiliano Todisco, Eurecom, FRANCE</div><div style=3D"font-size:1= 2.8px">=C2=A0 Hector Delgado, Eurecom, FRANCE</div><div style=3D"font-size:= 12.8px"><br></div><div style=3D"font-size:12.8px">CONTACT:</div><div style= =3D"font-size:12.8px"><br></div><div style=3D"font-size:12.8px">=C2=A0=C2= =A0<a href=3D"mailto:asvspoof2017@xxxxxxxx" target=3D"_blank">asvspoof2017= @xxxxxxxx</a></div><div style=3D"font-size:12.8px"><br></div><div style=3D= "font-size:12.8px"><br></div><div><br></div>-- <br><div class=3D"gmail_sign= ature"><div dir=3D"ltr"><div><div dir=3D"ltr"><div><div dir=3D"ltr"><div><d= iv dir=3D"ltr"><font face=3D"comic sans ms, sans-serif" size=3D"2" color=3D= "#000000">Dr. Md Sahidullah<br>Post-doctoral researcher<br>School of Comput= ing</font><div><font face=3D"comic sans ms, sans-serif" size=3D"2" color=3D= "#000000">University of Eastern Finland<br>Joensuu,=C2=A0=C2=A0P.O. Box 111= =C2=A0Finland-=C2=A080101</font></div><div><font face=3D"comic sans ms, san= s-serif" size=3D"2" color=3D"#000000">tel.=C2=A0+358-466250731<br>website:= =C2=A0<a href=3D"http://cs.joensuu.fi/~sahid/" target=3D"_blank">http://cs.= joensuu.fi/~sahid/</a></font></div></div></div></div></div></div></div></di= v></div> </div> --001a11372e92f55b540544586593--


This message came from the mail archive
/var/www/html/postings/2016/
maintained by:
DAn Ellis <dpwe@ee.columbia.edu>
Electrical Engineering Dept., Columbia University