[AUDITORY] ASVspoof 2019 CHALLENGE: Future horizons in spoofed/fake audio detection (Md Sahidullah )

Subject: [AUDITORY] ASVspoof 2019 CHALLENGE: Future horizons in spoofed/fake audio detection From: Md Sahidullah <sahidullahmd@xxxxxxxx> Date: Thu, 20 Dec 2018 11:50:20 +0100 List-Archive:<http://lists.mcgill.ca/scripts/wa.exe?LIST=AUDITORY> --0000000000004b9640057d71e624 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable [Apologies for possible cross-posting] **=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*= =3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D** *ASVspoof 2019 CHALLENGE:Future horizons in spoofed/fake audio detection**http://www.asvspoof.org/ <http://www.asvspoof.org/>* * =3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*= =3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D* Can you distinguish computer-generated or replayed speech from authentic/bona fide speech? Are you able to design algorithms to detect spoofs/fakes automatically? Are you concerned with the security of voice-driven interfaces? Are you searching for new challenges in machine learning and signal processing? *Join ASVspoof 2019* =E2=80=93 the effort to develop next-generation countermeasures for the automatic detection of spoofed/fake audio. In combining the forces of leading research institutes and industry, ASVspoof 2019 encompasses two separate sub-challenges in logical and physical access control, and provides a common database of the most advanced spoofing attacks to date. The aim is to study both the limits and opportunities of spoofing countermeasures in the context of automatic speaker verification and fake audio detection. *CHALLENGE TASK* Given a short audio clip, determine whether it represents authentic/bona fide human speech, or a spoof/fake (replay, synthesized speech or converted voice). You will be provided with a large database of labelled training and development data and will develop machine learning and signal processing countermeasures to distinguish automatically between the two. Countermeasure performance will be evaluated jointly with an automatic speaker verification (ASV) system provided by the organisers. *BACKGROUND:*The ASVspoof 2019 challenge follows on from two previous ASVspoof challenges, held in 2015 and 2017. The 2015 edition focused on spoofed speech generated with text-to-speech (TTS) and voice conversion (VC) technologies. The 2017 edition focused on replay spoofing. The 2019 edition is the first to address all three forms attack and the latest, cutting-edge spoofing attack technology. *ADVANCES:* Today=E2=80=99s state-of-the-art, TTS and VC technologies produce speech si= gnals that are as good as perceptually indistinguishable from bona fide speech. The LOGICAL ACCESS sub-challenge aims to determine whether the advances in TTS and VC pose a greater threat to the reliability of automatic speaker verification and spoofing countermeasure technologies. The PHYSICAL ACCESS sub-challenge builds upon the 2017 edition with a far more controlled evaluation setup which extends the focus of ASVspoof to fake audio detection in, e.g. the manipulation of voice-driven interfaces (smart speakers). *METRICS:* The 2019 edition also adopts a new metric, the tandem detection cost function (t-DCF). Adoption of the t-DCF metric aligns ASVspoof more closely to the field of ASV. The challenge nonetheless focuses on the development of standalone spoofing countermeasures; participation in ASVspoof 2019 does NOT require any expertise in ASV. The equal error rate (EER) used in previous editions remains as a secondary metric, supporting the wider implications of ASVspoof involving fake audio detection. *SCHEDULE:* Training and development data release: 19th December 2018 Evaluation data release: 15th February 2019 Deadline to submit evaluation scores: 22nd February 2019 Organisers return results to participants: 15th March 2019 INTERSPEECH paper submission deadline: 29th March 2019 *REGISTRATION:* Registration should be performed once only for each participating entity and by sending an email to registration@xxxxxxxx with =E2=80=98ASVspoof= 2019 registration=E2=80=99 as the subject line. The mail body should include: (i= ) the name of the team; (ii) the name of the contact person; (iii) their country; (iv) their status (academic/non-academic), and (v) the challenge scenario(s) for which they wish to participate (indicative only). Data download links will be communicated to registered contact persons only. *MAILING LIST:* Subscribe to general mailing list by sending e-mail with subject line =E2=80=98subscribe asvspoof2019=E2=80=99 to *sympa@xxxxxxxx <sympa@xxxxxxxx= poof.org>*. To post messages to the mailing list itself, send e-mails to *asvspoof2019@xxxxxxxx <asvspoof2019@xxxxxxxx>* *ORGANIZERS*:* Junichi Yamagishi, NII, Japan & Univ. of Edinburgh, UK Massimiliano Todisco, EURECOM, France Md Sahidullah, Inria, France H=C3=A9ctor Delgado, EURECOM, France Xin Wang, National Institute of Informatics, Japan Nicholas Evans, EURECOM, France Tomi Kinnunen, University of Eastern Finland, Finland Kong Aik Lee, NEC, JAPAN Ville Vestman, University of Eastern Finland, Finland (*) Equal contribution *CONTRIBUTORS:* University of Edinburgh, UK; Nara Institute of Science and Technology, Japan, University of Science and Technology of China, China; iFlytek Research, China; Saarland University / DFKI GmbH, Germany; Trinity College Dublin, Ireland; NTT Communication Science Laboratories, Japan; HOYA, Japan; Google LLC (Text-to-Speech team, Google Brain team, Deepmind); University of Avignon, France; Aalto University, Finland; University of Eastern Finland, Finland; EURECOM, France. *FURTHER INFORMATION:* *info@xxxxxxxx <info@xxxxxxxx>* --=20 Md Sahidullah website: *https://sites.google.com/site/iitkgpsahi/ <https://sites.google.com/site/iitkgpsahi/>* --0000000000004b9640057d71e624 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable <div dir=3D"ltr"><p class=3D"MsoNormal" style=3D"text-align:left;margin:0cm= 0cm 0.0001pt;line-height:normal;background-image:initial;background-positi= on:initial;background-size:initial;background-repeat:initial;background-ori= gin:initial;background-clip:initial;font-size:11pt;font-family:Calibri,sans= -serif">[Apologies for p= ossible cross-posting]<p class=3D"MsoNormal" style=3D"text-align= :left;margin:0cm 0cm 0.0001pt;line-height:normal;background-image:initial;b= ackground-position:initial;background-size:initial;background-repeat:initia= l;background-origin:initial;background-clip:initial;font-size:11pt;font-fam= ily:Calibri,sans-serif"> <p class=3D"MsoNormal" align=3D"center" = style=3D"margin:0cm 0cm 0.0001pt;text-align:center;line-height:normal;backg= round-image:initial;background-position:initial;background-size:initial;bac= kground-repeat:initial;background-origin:initial;background-clip:initial;fo= nt-size:11pt;font-family:Calibri,sans-serif">*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*= =3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D= *=3D*</b= ><p class=3D"MsoNormal" align=3D"center" style=3D"margin:0cm 0cm 0.0001= pt;text-align:center;line-height:normal;background-image:initial;background= -position:initial;background-size:initial;background-repeat:initial;backgro= und-origin:initial;background-clip:initial;font-size:11pt;font-family:Calib= ri,sans-serif">ASVspoof 20= 19 CHALLENGE: Future horizons in spoofed/fake audio detection = <a href= =3D"http://www.asvspoof.org/" target=3D"_blank" style=3D"color:blue">http:/= /www.asvspoof.org/</a>=C2=A0 =3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*= =3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D*=3D= *=3D*=3D Ca= n you distinguish computer-generated or replayed speech from authentic/bona= fide speech? Are you able to design algorithms to detect spoofs/fakes auto= matically?<p class=3D"MsoNormal" align=3D"center" style=3D"margi= n:0cm 0cm 0.0001pt;text-align:center;line-height:normal;background-image:in= itial;background-position:initial;background-size:initial;background-repeat= :initial;background-origin:initial;background-clip:initial;font-size:11pt;f= ont-family:Calibri,sans-serif">Are you concerned with the security of voice-driven interfaces?</p= ><p class=3D"MsoNormal" align=3D"center" style=3D"margin:0cm 0cm 0.0001pt;t= ext-align:center;line-height:normal;background-image:initial;background-pos= ition:initial;background-size:initial;background-repeat:initial;background-= origin:initial;background-clip:initial;font-size:11pt;font-family:Calibri,s= ans-serif">Are you searching = for new challenges in machine learning and signal processing?<p class=3D"MsoNormal" st= yle=3D"margin:0cm 0cm 0.0001pt;text-align:justify;line-height:normal;backgr= ound-image:initial;background-position:initial;background-size:initial;back= ground-repeat:initial;background-origin:initial;background-clip:initial;fon= t-size:11pt;font-family:Calibri,sans-serif">=C2=A0<= /span><p class=3D"MsoNormal" style=3D"margin:0cm 0cm 0.0001pt;text-alig= n:justify;line-height:normal;background-image:initial;background-position:i= nitial;background-size:initial;background-repeat:initial;background-origin:= initial;background-clip:initial;font-size:11pt;font-family:Calibri,sans-ser= if">Join ASVspoof 2019</sp= an>=C2=A0=E2=80=93 the ef= fort to develop next-generation countermeasures for the automatic detection= of spoofed/fake audio. In combining the forces of leading research institu= tes and industry, ASVspoof 2019 encompasses two separate sub-challenges in = logical and physical access control, and provides a common database of the = most advanced spoofing attacks to date. The aim is to study both the limits= and opportunities of spoofing countermeasures in the context of automatic = speaker verification and fake audio detection.<p class=3D"MsoNormal" style=3D"margin:0= cm 0cm 0.0001pt;text-align:justify;line-height:normal;background-image:init= ial;background-position:initial;background-size:initial;background-repeat:i= nitial;background-origin:initial;background-clip:initial;font-size:11pt;fon= t-family:Calibri,sans-serif">= CHALLENGE TASK<p class=3D"MsoNormal" style=3D"margin:0cm 0cm 0.0001p= t;text-align:justify;line-height:normal;background-image:initial;background= -position:initial;background-size:initial;background-repeat:initial;backgro= und-origin:initial;background-clip:initial;font-size:11pt;font-family:Calib= ri,sans-serif">Given a short = audio clip, determine whether it represents authentic/bona fide human speec= h, or a spoof/fake (replay, synthesized speech or converted voice). You wil= l be provided with a large database of labelled training and development da= ta and will develop machine learning and signal processing countermeasures = to distinguish automatically between the two. Countermeasure performance wi= ll be evaluated jointly with an automatic speaker verification (ASV) system= provided by the organisers.<p class=3D"MsoNormal" style=3D"marg= in:0cm 0cm 0.0001pt;text-align:justify;line-height:normal;background-image:= initial;background-position:initial;background-size:initial;background-repe= at:initial;background-origin:initial;background-clip:initial;font-size:11pt= ;font-family:Calibri,sans-serif">=C2=A0<p class=3D"MsoNormal" style=3D"margin:0cm 0cm 0.0001p= t;text-align:justify;line-height:normal;background-image:initial;background= -position:initial;background-size:initial;background-repeat:initial;backgro= und-origin:initial;background-clip:initial;font-size:11pt;font-family:Calib= ri,sans-serif">BACKGROUND:= The ASVspoof 2= 019 challenge follows on from two previous ASVspoof challenges, held in 201= 5 and 2017. The 2015 edition focused on spoofed speech generated with text-= to-speech (TTS) and voice conversion (VC) technologies. The 2017 edition fo= cused on replay spoofing. The 2019 edition is the first to address all thre= e forms attack and the latest, cutting-edge spoofing attack technology.</sp= an><p class=3D"Mso= Normal" style=3D"margin:0cm 0cm 0.0001pt;text-align:justify;line-height:nor= mal;background-image:initial;background-position:initial;background-size:in= itial;background-repeat:initial;background-origin:initial;background-clip:i= nitial;font-size:11pt;font-family:Calibri,sans-serif">=C2=A0<p class=3D"MsoNormal" style=3D"m= argin:0cm 0cm 0.0001pt;text-align:justify;line-height:normal;background-ima= ge:initial;background-position:initial;background-size:initial;background-r= epeat:initial;background-origin:initial;background-clip:initial;font-size:1= 1pt;font-family:Calibri,sans-serif">ADVANCES:<p class=3D"MsoNormal" style=3D"margin:0c= m 0cm 0.0001pt;text-align:justify;line-height:normal;background-image:initi= al;background-position:initial;background-size:initial;background-repeat:in= itial;background-origin:initial;background-clip:initial;font-size:11pt;font= -family:Calibri,sans-serif">T= oday=E2=80=99s state-of-the-art, TTS and VC technologies produce speech sig= nals that are as good as perceptually indistinguishable from bona fide spee= ch. The LOGICAL ACCESS sub-challenge aims to determine whether the advances= in TTS and VC pose a greater threat to the reliability of automatic speake= r verification and spoofing countermeasure technologies. The PHYSICAL ACCES= S sub-challenge builds upon the 2017 edition with a far more controlled eva= luation setup which extends the focus of ASVspoof to fake audio detection i= n, e.g. the manipulation of voice-driven interfaces (smart speakers).</span= ><p class=3D"MsoNormal" style=3D"margin:0cm 0cm 0.0001pt;text-align:jus= tify;line-height:normal;background-image:initial;background-position:initia= l;background-size:initial;background-repeat:initial;background-origin:initi= al;background-clip:initial;font-size:11pt;font-family:Calibri,sans-serif"><= span lang=3D"EN-GB" style=3D"font-size:12pt">=C2=A0<p class=3D"MsoNormal" style=3D"mar= gin:0cm 0cm 0.0001pt;text-align:justify;line-height:normal;background-image= :initial;background-position:initial;background-size:initial;background-rep= eat:initial;background-origin:initial;background-clip:initial;font-size:11p= t;font-family:Calibri,sans-serif">METRICS:<p class=3D"MsoNormal" style=3D"margin:0cm 0= cm 0.0001pt;text-align:justify;line-height:normal;background-image:initial;= background-position:initial;background-size:initial;background-repeat:initi= al;background-origin:initial;background-clip:initial;font-size:11pt;font-fa= mily:Calibri,sans-serif">The = 2019 edition also adopts a new metric, the tandem detection cost function (= t-DCF). Adoption of the t-DCF metric aligns ASVspoof more closely to the fi= eld of ASV. The challenge nonetheless focuses on the development of standal= one spoofing countermeasures; participation in ASVspoof 2019 does NOT requi= re any expertise in ASV. The equal error rate (EER) used in previous editio= ns remains as a secondary metric, supporting the wider implications of ASVs= poof involving fake audio detection. <p class=3D"MsoNormal" style=3D"margin:0cm 0cm 0.0001pt;t= ext-align:justify;line-height:normal;background-image:initial;background-po= sition:initial;background-size:initial;background-repeat:initial;background= -origin:initial;background-clip:initial;font-size:11pt;font-family:Calibri,= sans-serif">SCHEDULE:</spa= n><p class=3D"MsoNormal" style=3D"margin:0cm 0cm 0.0001pt 36pt;text= -align:justify;line-height:normal;background-image:initial;background-posit= ion:initial;background-size:initial;background-repeat:initial;background-or= igin:initial;background-clip:initial;font-size:11pt;font-family:Calibri,san= s-serif">Training and develop= ment data release: 19th December 2018<p class=3D"MsoNormal" style=3D"margin:0cm 0cm 0.= 0001pt 36pt;text-align:justify;line-height:normal;background-image:initial;= background-position:initial;background-size:initial;background-repeat:initi= al;background-origin:initial;background-clip:initial;font-size:11pt;font-fa= mily:Calibri,sans-serif">Eval= uation data release: 15th February 2019<p class=3D"MsoNormal" style=3D"margin:0cm 0cm = 0.0001pt 36pt;text-align:justify;line-height:normal;background-image:initia= l;background-position:initial;background-size:initial;background-repeat:ini= tial;background-origin:initial;background-clip:initial;font-size:11pt;font-= family:Calibri,sans-serif">De= adline to submit evaluation scores: 22nd February 2019<p class=3D"MsoNormal" style=3D"= margin:0cm 0cm 0.0001pt 36pt;text-align:justify;line-height:normal;backgrou= nd-image:initial;background-position:initial;background-size:initial;backgr= ound-repeat:initial;background-origin:initial;background-clip:initial;font-= size:11pt;font-family:Calibri,sans-serif">Organisers return results to participants: 15th March 2019</sp= an><p class=3D"Mso= Normal" style=3D"margin:0cm 0cm 0.0001pt 36pt;text-align:justify;line-heigh= t:normal;background-image:initial;background-position:initial;background-si= ze:initial;background-repeat:initial;background-origin:initial;background-c= lip:initial;font-size:11pt;font-family:Calibri,sans-serif">INTERSPEECH paper submission deadline: 29th M= arch 2019<p= class=3D"MsoNormal" style=3D"margin:0cm 0cm 0.0001pt;text-align:justify;li= ne-height:normal;background-image:initial;background-position:initial;backg= round-size:initial;background-repeat:initial;background-origin:initial;back= ground-clip:initial;font-size:11pt;font-family:Calibri,sans-serif">=C2=A0<p class=3D"MsoNorma= l" style=3D"margin:0cm 0cm 0.0001pt;text-align:justify;line-height:normal;b= ackground-image:initial;background-position:initial;background-size:initial= ;background-repeat:initial;background-origin:initial;background-clip:initia= l;font-size:11pt;font-family:Calibri,sans-serif">REGISTRATION:<p class=3D"MsoNormal" style=3D"margin:= 0cm 0cm 0.0001pt;text-align:justify;line-height:normal;background-image:ini= tial;background-position:initial;background-size:initial;background-repeat:= initial;background-origin:initial;background-clip:initial;font-size:11pt;fo= nt-family:Calibri,sans-serif">Registration should be performed once only for each participating entity a= nd by sending an email to=C2=A0= <a href=3D"mailto:registration@xxxxxxxx" target=3D"_blank" style=3D"col= or:blue">registration@xxxxxxxx</a>=C2=A0with =E2=80=98ASVspoof 2019 registration=E2=80= =99 as the subject line. The mail body should include: (i)=C2=A0the name of= the team; (ii)=C2=A0the name of the contact person; (iii)=C2=A0their count= ry; (iv)=C2=A0their status (academic/non-academic), and (v)=C2=A0the challe= nge scenario(s) for which they wish to participate (indicative only). Data = download links will be communicated to registered contact persons only.</sp= an><p class=3D"Mso= Normal" style=3D"margin:0cm 0cm 0.0001pt;text-align:justify;line-height:nor= mal;background-image:initial;background-position:initial;background-size:in= itial;background-repeat:initial;background-origin:initial;background-clip:i= nitial;font-size:11pt;font-family:Calibri,sans-serif">=C2=A0<p class=3D"MsoNormal" style=3D"margin:0cm 0cm 0.0001pt= ;text-align:justify;line-height:normal;background-image:initial;background-= position:initial;background-size:initial;background-repeat:initial;backgrou= nd-origin:initial;background-clip:initial;font-size:11pt;font-family:Calibr= i,sans-serif">MAILING LIST= :</p= ><p class=3D"MsoNormal" style=3D"margin:0cm 0cm 0.0001pt;text-align:justify= ;line-height:normal;background-image:initial;background-position:initial;ba= ckground-size:initial;background-repeat:initial;background-origin:initial;b= ackground-clip:initial;font-size:11pt;font-family:Calibri,sans-serif"><span= lang=3D"EN-GB" style=3D"font-size:12pt">Subscribe to general mailing list = by sending e-mail with subject line =E2=80=98subscribe asvspoof2019=E2=80= =99 to=C2=A0= <a href=3D"mailto:= sympa@xxxxxxxx" target=3D"_blank" style=3D"color:blue">sympa@xxxxxxxx= rg</a>. To post me= ssages to the mailing list itself, send e-mails to=C2=A0<a href=3D"mailto:asvspoof2= 019@xxxxxxxx" target=3D"_blank" style=3D"color:blue">asvspoof2019@xxxxxxxx= oof.org</a>= <p class=3D"MsoNormal" style=3D"margin:0cm 0cm 0.0001pt;text-align:just= ify;line-height:normal;background-image:initial;background-position:initial= ;background-size:initial;background-repeat:initial;background-origin:initia= l;background-clip:initial;font-size:11pt;font-family:Calibri,sans-serif"><s= pan lang=3D"EN-GB" style=3D"font-size:12pt">=C2=A0<p class=3D"MsoNormal" style=3D"marg= in:0cm 0cm 0.0001pt;line-height:normal;background-image:initial;background-= position:initial;background-size:initial;background-repeat:initial;backgrou= nd-origin:initial;background-clip:initial;font-size:11pt;font-family:Calibr= i,sans-serif">ORGANIZERS<s= up>*: <br= >Junichi Yamagishi, NII, Japan & Univ. of Edinburgh, UK<p class=3D"MsoNormal" styl= e=3D"margin:0cm 0cm 0.0001pt;text-align:justify;line-height:normal;backgrou= nd-image:initial;background-position:initial;background-size:initial;backgr= ound-repeat:initial;background-origin:initial;background-clip:initial;font-= size:11pt;font-family:Calibri,sans-serif">Massimiliano Todisco, EURECOM, France<p class=3D"MsoNormal" style=3D"marg= in:0cm 0cm 0.0001pt;text-align:justify;line-height:normal;background-image:= initial;background-position:initial;background-size:initial;background-repe= at:initial;background-origin:initial;background-clip:initial;font-size:11pt= ;font-family:Calibri,sans-serif">Md Sahidullah, Inria, France<p class=3D"MsoNormal" style=3D"margin:0cm 0cm 0.0001p= t;text-align:justify;line-height:normal;background-image:initial;background= -position:initial;background-size:initial;background-repeat:initial;backgro= und-origin:initial;background-clip:initial;font-size:11pt;font-family:Calib= ri,sans-serif">H=C3=A9ctor De= lgado, EURECOM, France= <p class=3D"MsoNormal" style=3D"margin:0cm 0cm 0.0001pt;text-ali= gn:justify;line-height:normal;background-image:initial;background-position:= initial;background-size:initial;background-repeat:initial;background-origin= :initial;background-clip:initial;font-size:11pt;font-family:Calibri,sans-se= rif">Xin Wang, National Insti= tute of Informatics, Japan<p class=3D"MsoNormal" style=3D"margin:0cm 0cm 0.0001pt;text= -align:justify;line-height:normal;background-image:initial;background-posit= ion:initial;background-size:initial;background-repeat:initial;background-or= igin:initial;background-clip:initial;font-size:11pt;font-family:Calibri,san= s-serif">Nicholas Evans, EURE= COM, France= <p class=3D"MsoNormal" style=3D"margin:0cm 0cm 0.0001pt;text-align:justify;= line-height:normal;background-image:initial;background-position:initial;bac= kground-size:initial;background-repeat:initial;background-origin:initial;ba= ckground-clip:initial;font-size:11pt;font-family:Calibri,sans-serif">Tomi Kinnunen, University of Easter= n Finland, Finland</sp= an><p class=3D"MsoNormal" style=3D"margin:0cm 0cm 0.0001pt;text-align:j= ustify;line-height:normal;background-image:initial;background-position:init= ial;background-size:initial;background-repeat:initial;background-origin:ini= tial;background-clip:initial;font-size:11pt;font-family:Calibri,sans-serif"= >Kong Aik Lee, NEC, JAPAN</sp= an><p class=3D"Mso= Normal" style=3D"margin:0cm 0cm 12pt;text-align:justify;line-height:normal;= background-image:initial;background-position:initial;background-size:initia= l;background-repeat:initial;background-origin:initial;background-clip:initi= al;font-size:11pt;font-family:Calibri,sans-serif">Ville Vestman, University of Eastern Finland, Finland<= /span><p class=3D"MsoNormal" style=3D"margin:0cm 0cm 12pt;text-align:ju= stify;line-height:normal;background-image:initial;background-position:initi= al;background-size:initial;background-repeat:initial;background-origin:init= ial;background-clip:initial;font-size:11pt;font-family:Calibri,sans-serif">= (*) Equal contribution= <p class=3D"MsoNormal" style=3D"margin:0cm 0cm 0.0001pt;text-align:just= ify;line-height:normal;background-image:initial;background-position:initial= ;background-size:initial;background-repeat:initial;background-origin:initia= l;background-clip:initial;font-size:11pt;font-family:Calibri,sans-serif"><b= >CONTRIBUTORS:<= span lang=3D"EN-GB" style=3D"font-size:12pt"><p class=3D"Mso= Normal" style=3D"margin:0cm 0cm 0.0001pt;text-align:justify;line-height:nor= mal;background-image:initial;background-position:initial;background-size:in= itial;background-repeat:initial;background-origin:initial;background-clip:i= nitial;font-size:11pt;font-family:Calibri,sans-serif">University of Edinburgh, UK; Nara Institute of Sci= ence and Technology, Japan, University of Science and Technology of China, = China; iFlytek Research, China; Saarland University / DFKI GmbH, Germany; T= rinity College Dublin, Ireland; NTT Communication Science Laboratories, Jap= an; HOYA, Japan;=C2=A0Google LLC (Text-to-Speech team, Google Brain team, D= eepmind); University of Avignon, France; Aalto University, Finland; Univers= ity of Eastern Finland, Finland; EURECOM, France.<p class=3D"MsoNormal" style=3D"margi= n:0cm 0cm 0.0001pt;text-align:justify;line-height:normal;background-image:i= nitial;background-position:initial;background-size:initial;background-repea= t:initial;background-origin:initial;background-clip:initial;font-size:11pt;= font-family:Calibri,sans-serif"> FURTHER INFORMATION:<p class=3D"MsoNormal" style=3D"margin:0cm 0c= m 0.0001pt;text-align:justify;line-height:normal;background-image:initial;b= ackground-position:initial;background-size:initial;background-repeat:initia= l;background-origin:initial;background-clip:initial;font-size:11pt;font-fam= ily:Calibri,sans-serif"><a href=3D"mailto:info@xxxxxxxx" target=3D"_blank" style=3D"co= lor:blue">info@xxxxxxxx</a><div class=3D"gmail-yj6qo gma= il-ajU" style=3D"outline:none;padding:10px 0px;width:22px;margin:2px 0px 0p= x"> </div><div> </div>-- <b= r><div dir=3D"ltr" class=3D"gmail_signature" data-smartmail=3D"gmail_signat= ure"><div dir=3D"ltr"><div><div dir=3D"ltr"><div><div dir=3D"ltr"><div><div= dir=3D"ltr"><div dir=3D"ltr"><div dir=3D"ltr"><div dir=3D"ltr">Md Sahidullah</= font><div>website:=C2=A0<a href=3D"https://sites.google.com/site/iitkgpsahi= /" target=3D"_blank">https://sites.google.com/site/iitkgpsahi/</a></fon= t></div></div></div></div></div></div></div></div></div></div></div></div><= /div> --0000000000004b9640057d71e624--

This message came from the mail archive
src/postings/2018/
maintained by:

DAn Ellis <dpwe@ee.columbia.edu>
Electrical Engineering Dept., Columbia University