Subject: [AUDITORY] ICASSP 2021 Acoustic Echo Cancellation Challenge From: Ross Cutler <000001144701df79-dmarc-request@xxxxxxxx> Date: Thu, 6 Aug 2020 05:15:02 +0000--_000_CO2PR00MB0183CF086843A7937C4DD636EB480CO2PR00MB0183namp_ Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Our team at Microsoft recently organized the Deep Noise Suppression Challen= ge at INTERSPEECH 2020<https://dns-challenge.azurewebsites.net/> to help st= imulate research in this important area of research. The response from that= challenge exceeded our expectations with 19 teams submitting 28 models for= noise suppression. The dataset and test framework continue to be cloned an= d forked at a good rate since the challenge, indicating it has good utility= for the research community. Another important area for speech enhancement is acoustic echo cancellation= , which is still a top issue in audio communication and conferencing system= s. We are submitting a proposal to organize an Acoustic Echo Cancellation C= hallenge at ICASSP 2021 to stimulate research in this area. The challenge w= ill have an open source dataset for acoustic echo cancellation and detectio= n, captured using a large-scale crowdsourcing effort. We will provide a dat= aset of 10,000 different real environments, audio devices, and human speake= rs in the following scenarios: 1. Far-end single talk, no echo path change 2. Far-end single talk, echo path change 3. Near-end single talk, no echo path change 4. Double talk, no echo path change 5. Double talk, echo path change 6. Sweep signal for RT60 estimation Each scenario includes the speaker, microphone, and loopback signal. Beside= s the real data, we will also provide synthetic data which is useful for tr= aining. We propose two tracks: 1. Real-time acoustic echo detection (AED) classifier. 2. Real-time acoustic echo cancellation (AEC). In both the tracks: * The AED/AEC must run in less than T/2 (in ms) to process a frame of s= ize T (in ms) on an Intel Core i5 quad-core machine clocked at 2.4 GHz or e= quivalent processors. Frame length T must be less than or equal to 40ms (0<= T<=3D40ms). * The AED/AEC method may have a maximum of 40ms look ahead. To infer th= e current frame T (in ms), the algorithm can access any number of past fram= es but only 40ms of future frames (T+40ms). * We will provide a base model implementation. * We will provide an objective and subjective evaluation framework. * The AED/AEC can be a deep model, a traditional signal processing algo= rithm, or a mix of the two. There are no restrictions on the AED/AEC aside = from the run time and receptive field described above. Participants can enter either one or both tracks. Prizes will be awarded to= the top three participants in each track. For the proposal, ICASSP requires us to provide them the list of potential = participants with their potential title and author(s) details. If you are i= nterested to participate in this challenge, please send an email to ross.cu= tler@xxxxxxxx the following details by August 9th. 1. A potential title of your paper. This can be updated later. 2. A potential list of authors. This can be updated later. 3. A webpage/URL of your personal webpage, LinkedIn page, or your resear= ch lab page. We look forward to hearing back from you and we hope to see you participate= in the challenge. Regards, Acoustic Echo Cancellation Challenge Organizers --_000_CO2PR00MB0183CF086843A7937C4DD636EB480CO2PR00MB0183namp_ Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable <html> <head> <meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3Dus-ascii"= > </head> <body> <div dir=3D"ltr"> <div></div> <div> <div dir=3D"ltr"></div> </div> <div id=3D"id-b452c58a-6663-4f41-9338-4b36edd800da" class=3D"ms-outlook-mob= ile-reference-message"> <meta content=3D"text/html; charset=3Dus-ascii"> <meta name=3D"Generator" content=3D"Microsoft Word 15 (filtered medium)"> <style>=0A= <!--=0A= @xxxxxxxx=0A= {font-family:Wingdings}=0A= @xxxxxxxx=0A= {font-family:"Cambria Math"}=0A= @xxxxxxxx=0A= {font-family:Calibri}=0A= p.MsoNormal, li.MsoNormal, div.MsoNormal=0A= {margin-top:0in;=0A= margin-right:0in;=0A= margin-bottom:8.0pt;=0A= margin-left:0in;=0A= line-height:106%;=0A= font-size:11.0pt;=0A= font-family:"Calibri",sans-serif}=0A= a:link, span.MsoHyperlink=0A= {color:#0563C1;=0A= text-decoration:underline}=0A= p.MsoListParagraph, li.MsoListParagraph, div.MsoListParagraph=0A= {margin-top:0in;=0A= margin-right:0in;=0A= margin-bottom:8.0pt;=0A= margin-left:.5in;=0A= line-height:106%;=0A= font-size:11.0pt;=0A= font-family:"Calibri",sans-serif}=0A= p.MsoListParagraphCxSpFirst, li.MsoListParagraphCxSpFirst, div.MsoListParag= raphCxSpFirst=0A= {margin-top:0in;=0A= margin-right:0in;=0A= margin-bottom:0in;=0A= margin-left:.5in;=0A= line-height:106%;=0A= font-size:11.0pt;=0A= font-family:"Calibri",sans-serif}=0A= p.MsoListParagraphCxSpMiddle, li.MsoListParagraphCxSpMiddle, div.MsoListPar= agraphCxSpMiddle=0A= {margin-top:0in;=0A= margin-right:0in;=0A= margin-bottom:0in;=0A= margin-left:.5in;=0A= line-height:106%;=0A= font-size:11.0pt;=0A= font-family:"Calibri",sans-serif}=0A= p.MsoListParagraphCxSpLast, li.MsoListParagraphCxSpLast, div.MsoListParagra= phCxSpLast=0A= {margin-top:0in;=0A= margin-right:0in;=0A= margin-bottom:8.0pt;=0A= margin-left:.5in;=0A= line-height:106%;=0A= font-size:11.0pt;=0A= font-family:"Calibri",sans-serif}=0A= span.EmailStyle17=0A= {font-family:"Calibri",sans-serif;=0A= color:windowtext}=0A= .MsoChpDefault=0A= {font-family:"Calibri",sans-serif}=0A= @xxxxxxxx WordSection1=0A= {margin:1.0in 1.0in 1.0in 1.0in}=0A= div.WordSection1=0A= {}=0A= ol=0A= {margin-bottom:0in}=0A= ul=0A= {margin-bottom:0in}=0A= -->=0A= </style> <div class=3D"WordSection1"> <p class=3D"MsoNormal">Our team at Microsoft recently organized the <a href= =3D"https://dns-challenge.azurewebsites.net/"> Deep Noise Suppression Challenge at INTERSPEECH 2020</a> to help stimulate = research in this important area of research. The response from that challen= ge exceeded our expectations with 19 teams submitting 28 models for noise s= uppression. The dataset and test framework continue to be cloned and forked at a good rate since the challe= nge, indicating it has good utility for the research community.</p> <p class=3D"MsoNormal">Another important area for speech enhancement is aco= ustic echo cancellation, which is still a top issue in audio communication = and conferencing systems. We are submitting a proposal to organize an Acous= tic Echo Cancellation Challenge at ICASSP 2021 to stimulate research in this area. The challenge will have an= open source dataset for acoustic echo cancellation and detection, captured= using a large-scale crowdsourcing effort. We will provide a dataset of 10,= 000 different real environments, audio devices, and human speakers in the following scenarios:</p> <ol start=3D"1" type=3D"1" style=3D"margin-top:0in"> <li class=3D"MsoListParagraphCxSpFirst" style=3D"margin-left:0in; mso-list:= l3 level1 lfo1"> Far-end single talk, no echo path change</li><li class=3D"MsoListParagraphC= xSpMiddle" style=3D"margin-left:0in; mso-list:l3 level1 lfo1"> Far-end single talk, echo path change</li><li class=3D"MsoListParagraphCxSp= Middle" style=3D"margin-left:0in; mso-list:l3 level1 lfo1"> Near-end single talk, no echo path change</li><li class=3D"MsoListParagraph= CxSpMiddle" style=3D"margin-left:0in; mso-list:l3 level1 lfo1"> Double talk, no echo path change</li><li class=3D"MsoListParagraphCxSpMiddl= e" style=3D"margin-left:0in; mso-list:l3 level1 lfo1"> Double talk, echo path change</li><li class=3D"MsoListParagraphCxSpLast" st= yle=3D"margin-left:0in; mso-list:l3 level1 lfo1"> Sweep signal for RT60 estimation</li></ol> <p class=3D"MsoNormal">Each scenario includes the speaker, microphone, and = loopback signal. Besides the real data, we will also provide synthetic data= which is useful for training.</p> <p class=3D"MsoNormal">We propose two tracks:</p> <ol start=3D"1" type=3D"1" style=3D"margin-top:0in"> <li class=3D"MsoListParagraphCxSpFirst" style=3D"margin-left:0in; mso-list:= l2 level1 lfo2"> Real-time acoustic echo detection (AED) classifier. </li><li class=3D"MsoLi= stParagraphCxSpLast" style=3D"margin-left:0in; mso-list:l2 level1 lfo2"> Real-time acoustic echo cancellation (AEC). </li></ol> <p class=3D"MsoNormal">In both the tracks:</p> <ul type=3D"disc" style=3D"margin-top:0in"> <li class=3D"MsoListParagraphCxSpFirst" style=3D"margin-left:0in; mso-list:= l0 level1 lfo3"> The AED/AEC must run in less than T/2 (in ms) to process a frame of size T = (in ms) on an Intel Core i5 quad-core machine clocked at 2.4 GHz or equival= ent processors. Frame length T must be less than or equal to 40ms (0<T&l= t;=3D40ms).</li><li class=3D"MsoListParagraphCxSpMiddle" style=3D"margin-le= ft:0in; mso-list:l0 level1 lfo3"> The AED/AEC method may have a maximum of 40ms look ahead. To infer the curr= ent frame T (in ms), the algorithm can access any number of past frames but= only 40ms of future frames (T+40ms).</li><li class=3D"MsoListParagraphCxSp= Middle" style=3D"margin-left:0in; mso-list:l0 level1 lfo3"> We will provide a base model implementation.</li><li class=3D"MsoListParagr= aphCxSpMiddle" style=3D"margin-left:0in; mso-list:l0 level1 lfo3"> We will provide an objective and subjective evaluation framework.</li><li c= lass=3D"MsoListParagraphCxSpLast" style=3D"margin-left:0in; mso-list:l0 lev= el1 lfo3"> The AED/AEC can be a deep model, a traditional signal processing algorithm,= or a mix of the two. There are no restrictions on the AED/AEC aside from t= he run time and receptive field described above.</li></ul> <p class=3D"MsoNormal">Participants can enter either one or both tracks. Pr= izes will be awarded to the top three participants in each track.</p> <p class=3D"MsoNormal">For the proposal, ICASSP requires us to provide them= the list of potential participants with their potential title and author(s= ) details. If you are interested to participate in this challenge, please s= end an email to <b>ross.cutler@xxxxxxxx</b> the following details by <b>August 9th</b>= .</p> <ol start=3D"1" type=3D"1" style=3D"margin-top:0in"> <li class=3D"MsoListParagraphCxSpFirst" style=3D"margin-left:0in; mso-list:= l1 level1 lfo4"> A potential title of your paper. This can be updated later.</li><li class= =3D"MsoListParagraphCxSpMiddle" style=3D"margin-left:0in; mso-list:l1 level= 1 lfo4"> A potential list of authors. This can be updated later.</li><li class=3D"Ms= oListParagraphCxSpLast" style=3D"margin-left:0in; mso-list:l1 level1 lfo4"> A webpage/URL of your personal webpage, LinkedIn page, or your research lab= page.</li></ol> <p class=3D"MsoNormal">We look forward to hearing back from you and we hope= to see you participate in the challenge.</p> <p class=3D"MsoNormal">Regards,</p> <p class=3D"MsoNormal">Acoustic Echo Cancellation Challenge Organizers</p> <p class=3D"MsoNormal"> </p> </div> </div> </div> </body> </html> --_000_CO2PR00MB0183CF086843A7937C4DD636EB480CO2PR00MB0183namp_--