[AUDITORY] Automated audio captioning task at DCASE2021 (Konstantinos Drosos)

Subject: [AUDITORY] Automated audio captioning task at DCASE2021 From: Konstantinos Drosos <Konstantinos Drosos> Date: Wed, 10 Mar 2021 11:45:34 +0200 --Apple-Mail=_D5B30232-7486-499A-AF76-33E89EB00709 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=utf-8 Dear all,=20 =E2=80=94 Apologies for cross-posting =E2=80=94=20 I=E2=80=99m happy to share that the task of automated audio captioning = will be hosted in DCASE, again this year! Check the task web page at: http://dcase.community/challenge2021/task-automatic-audio-captioning = <http://dcase.community/challenge2021/task-automatic-audio-captioning> There are also some exciting changes!=20 This year the participants are allowed to use any external data, models, = or methods that they want to, pushing further the limits and the = performance of automated audio captioning methods. For example, you can = use other automated audio captioning datasets, sound event datasets, or = even natural language processing datasets. But you can also you = pre-trained models and methods, like BERT, YAMNet, and others!=20 We have compiled a GitHub repo that hosts some of the suggestions that = can get you started:=20 https://github.com/audio-captioning/audio-captioning-resources = <https://github.com/audio-captioning/audio-captioning-resources> Also, we are now finalizing the release of more data for Clotho (an = increase of around 40%), which can be used for developing your automated = audio captioning methods!=20 Visit the website of the task at DCASE2021 and stay tuned for more! Important dates:=20 Challenge deadline: 15th of June, 2021 Challenge results: 01st - 07th of July, 2021 Visit the DCASE page for more: http://dcase.community = <http://dcase.community/> Enjoy!=20 On behalf of the organizers of the automated audio captioning task,=20 # Kostas --Apple-Mail=_D5B30232-7486-499A-AF76-33E89EB00709 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=utf-8 <html><head><meta http-equiv=3D"Content-Type" content=3D"text/html; = charset=3Dutf-8"></head><body style=3D"word-wrap: break-word; = -webkit-nbsp-mode: space; line-break: after-white-space;" class=3D""><span= style=3D"caret-color: rgb(0, 0, 0); color: rgb(0, 0, 0);" class=3D"">Dear= all,  =E2=80=94 Apologies for cross-posting = =E2=80=94  I=E2=80=99m happy to share that the = task of automated audio captioning will be hosted in DCASE, again this = year! Check the task web page at: <a = href=3D"http://dcase.community/challenge2021/task-automatic-audio-captioni= ng" = class=3D"">http://dcase.community/challenge2021/task-automatic-audio-capti= oning</a> There are also some exciting changes!  <span= style=3D"caret-color: rgb(0, 0, 0); color: rgb(0, 0, 0);" class=3D"">This= year the participants are allowed to use any external data, models, or = methods that they want to, pushing further the limits and the = performance of automated audio captioning methods. For example, you can = use other automated audio captioning datasets, sound event datasets, or = even natural language processing datasets. But you can also you = pre-trained models and methods, like BERT, YAMNet, and = others!  We have compiled a GitHub repo that = hosts some of the suggestions that can get you started:  <a = href=3D"https://github.com/audio-captioning/audio-captioning-resources" = class=3D"">https://github.com/audio-captioning/audio-captioning-resources<= /a> Also, we are now finalizing the release of more data for = Clotho (an increase of around 40%), which can be used for developing = your automated audio captioning methods!  <span= style=3D"caret-color: rgb(0, 0, 0); color: rgb(0, 0, 0);" = class=3D"">Visit the website of the task at DCASE2021 and stay tuned for = more! Important dates:  Challenge deadline: 15th = of June, 2021 Challenge results: 01st - 07th of July, = 2021 Visit the DCASE page for more: <a = href=3D"http://dcase.community" class=3D"">http://dcase.community</a> <span= style=3D"caret-color: rgb(0, 0, 0); color: rgb(0, 0, 0);" = class=3D"">Enjoy!  On behalf of the organizers of the = automated audio captioning task, <div class=3D""> <div class=3D""> <div># Kostas</div> </div> </div></body></html>= --Apple-Mail=_D5B30232-7486-499A-AF76-33E89EB00709--

This message came from the mail archive
src/postings/2021/
maintained by:

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