@tom_kenter@maston.social (@TomKenter) header
@tom_kenter@maston.social (@TomKenter) profile picture

@[email protected]

@TomKenter

Research at Google UK on NLU and TTS. https://nitter.gq/t.co/xQtZY3oYSQ… I also 🎸 music: https://nitter.gq/t.co/PNo4S482w4…

 Amsterdam
tomkenter.nl
Joined September 2011
  • Tweets587
  • Following1.3K
  • Followers962
In the debates on double or single blind reviewing and arXiv, I remember this excellent study, which concluded "single-blind reviewing confers a significant advantage to papers with famous authors and authors from high-prestige institutions" pnas.org/doi/10.1073/pnas.17…
Reviewer bias in single- versus double-blind peer review | PNAS

Reviewer bias in single- versus double-blind peer review | PNAS

Peer review may be “single-blind,” in which reviewers are aware of the names and affiliations of paper authors, or “double-blind,” in which this in...

pnas.org
(((ل()(ل() 'yoav))))👾 (@yoavgo) profile picture(((ل()(ل() 'yoav))))👾
@yoavgo9 Sep 2023
a few more thoughts about the politics of ACL, anon policy, desire for diversity, etc.
3
5
2,272
Bard is live now. Sign up, use it, and help us keep improving it at bard.google.com. nitter.gq/sundarpichai/status/16…
Sundar Pichai (@sundarpichai) profile pictureSundar Pichai
@sundarpichai21 Mar 2023
We're expanding access to Bard in US + UK with more countries ahead, it's an early experiment that lets you collaborate with generative AI. Hope Bard sparks more creativity and curiosity, and will get better with feedback. Sign up: bard.google.com blog.google/technology/ai/tr…
Try Bard and share your feedback

Try Bard and share your feedback

Bard is an early experiment from Google that lets you collaborate with generative AI. If you’re in the U.S. and U.K., you can sign up to try it.

blog.google
1
8
53
10,162
Good news: the timeline has shifted so reviewers have more time for reviews. There will be emails sent out, but the new deadline for reviewing is ***Dec 27th.***
Amanda Stent (@astent) profile pictureAmanda Stent@astent30 Nov 2021
Replying to @ReviewAcl
Breaking news: The @aclmeeting program chairs have pushed out the ACL 2022 commitment date to Jan 15, thus allowing for a reviewing period of 11/29-12/27. Updated timeline at aclrollingreview.org/modifie…. Emails will go out to AEs and reviewers.
Modified Timeline November

Modified Timeline November

For November, working with the ACL 2022 senior area chairs and program chairs, we will have a slightly modified ARR timeline. * Nov 19-22 - SACs check AE assignments * Nov 25 - finalized AE assignm...

aclrollingreview.org
3
11
3
46
So exciting! 😃 The #igluContest has officially started! If you are interested in interactive grounded language learning, definitely consider submitting! We have two tracks, check them out below 👇 #NeurIPS2021
iglu-contest @ NeurIPS (@IgluContest) profile pictureiglu-contest @ NeurIPS@IgluContest30 Jul 2021
We are happy to announce that the submission system is up and running! So #igluContest has officially started: Architect Task: competitions.codalab.org/com… Builder Task: competitions.codalab.org/com… Please register and try it out! #NeurIPS2021
4
19
We're organizing a workshop on personal knowledge graphs at the #AKBC2021 conference. Submissions are due Sep 6. Please help spread the word pkgs.ws/
19
26
Hard to believe it’s finally here but it's true. My debut album, Child of the State, is out NOW!!! Listen loud. ayron.lnk.to/childofthestate…
35
58
30
289
I know Twitter is meant to be used for lamenting the state of reviewing in NLP. But can we also appreciate how cool it is to have a job that involves spending a little time every week helping anonymous strangers improve their research?
3
8
2
206
New paper from our team: Isaac Elias, Heiga Zen, Jonathan Shen, Yu Zhang, Ye Jia, RJ. Ryan, Yonghui Wu "Parallel Tacotron 2: A Non-Autoregressive Neural TTS Model with Differentiable Duration Modeling" Arxiv: arxiv.org/abs/2103.14574 Samples: google.github.io/tacotron/pu…
2
29
3
129
New paper from our team: Ye Jia, Heiga Zen, Jonathan Shen, Yu Zhang, Yonghui Wu "PnG BERT: Augmented BERT on Phonemes and Graphemes for Neural TTS" Arxiv: arxiv.org/abs/2103.15060 Samples: google.github.io/tacotron/pu…
1
25
97
``Parallel Tacotron 2: A Non-Autoregressive Neural TTS Model with Differentiable Duration Modeling. (arXiv:2103.14574v1 [cs.SD]),'' Isaac Elias, Heiga Zen, Jonathan Shen, Yu Zhang, Jia Ye, RJ Ryan, Yonghui Wu, ift.tt/2PxQDeq
7
25
New work with Ron Weiss, @EricBattenberg, @sorooshooryad, @dpkingma -- finally achieving what @log_pie and I set out to do in 2016 before switching to spectrograms: direct waveform generation from characters. (1/7) abs: arxiv.org/abs/2011.03568 samples: google.github.io/tacotron/pu…
Wave-Tacotron: Spectrogram-free end-to-end text-to-speech synthesis

Wave-Tacotron: Spectrogram-free end-to-end text-to-speech synthesis

We describe a sequence-to-sequence neural network which directly generates speech waveforms from text inputs. The architecture extends the Tacotron model by incorporating a normalizing flow into...

arxiv.org
AK (@_akhaliq) profile pictureAK
@_akhaliq10 Nov 2020
Wave-Tacotron: Spectrogram-free end-to-end text-to-speech synthesis pdf: arxiv.org/pdf/2011.03568.pdf abs: arxiv.org/abs/2011.03568 project page: google.github.io/tacotron/pu…
Image from AK (@_akhaliq)
1
33
2
88
I have a PhD position in #conversational #AI (partially funded by @GoogleAI). Please help RT krisztianbalog.com/2020-11-0…
PhD position in Conversational AI

PhD position in Conversational AI

I have a fully-funded PhD position in Conversational AI. This position is partially funded by an unrestricted gift from Google, and will be co-supervised by Google research scientist Filip Radlinsk…

krisztianbalog.com
1
40
2
59
There are models to identify the language used in a piece of written text which can recognize 1000+ languages. What happens if you apply them to as many websites as possible? Turns out, you have to strengthen the classifier by quite a bit to get usable results. But when you do...
arXiv CS-CL (@arxiv_cscl) profile picturearXiv CS-CL@arxiv_cscl29 Oct 2020
Language ID in the Wild: Unexpected Challenges on the Path to a Thousand-Language Web Text Corpus arxiv.org/abs/2010.14571
2
16
5
50
Hey y'all? If a paper has been uploaded to ArXiv and also published in a peer reviewed venue, the correct citation is the *peer reviewed version*. Don't just copy paste the default Google scholar BibTex citation. Check your citations & make sure the paper provenance is correct.
11
26
161
The code for our #SIGIR2020 paper is now publicly available: Policy-Aware Unbiased Learning to Rank for Top-k Rankings github.com/HarrieO/2020topku… The code can be used to replicate our experiments, the readme includes some examples to get you started.
2
4
School "buses" like this would be awesome! Health benefits, a bit of excess energy burned off before arriving at school, chance to all work together on getting to school, great for the environment, ... Probably way cheaper than a fuel-burning school bus, too.
Brent Toderian (@BrentToderian) profile pictureBrent Toderian
@BrentToderian7 Sep 2020
These Dutch “bicycle buses” are transporting 50+ kids getting the kind of exercise that leads to better learning, who DON’T need to be driven & dropped off by parents adding lots of local car traffic. Nijmegen, NL video via @MartijnLi HT @urbanthoughts11
24
180
24
1,207
My first (two) papers in Text-To-Speech (TTS) accepted at #interspeech2020! (w/ @TomKenter and Rob Clark) 1. StrawNet improves WaveNet training for low resource TTS google.github.io/StrawNet/ 2. CHiVE-BERT improves English TTS prosody google.github.io/chive-proso…
1
12
28
#CIKM2020 full paper accepted: When Inverse Propensity Scoring does not Work: Affine Corrections for Unbiased Learning to Rank Joint work with @AliVardasbi and @mdr We show that IPS cannot correct for trust-bias and introduce a novel estimator for counterfactual LTR that can.
2
6
49
This paper is good and you should read it (@samiraabnar & @wzuidema)
Carolin Lawrence (@caro__lawrence) profile pictureCarolin Lawrence@caro__lawrence16 Jul 2020
Replying to @Tim_Dettmers
There is an ACL 2020 paper that offers two post-hoc interpretability methods that traces back what the attention means for each token: Quantifying Atention Flow in Transformers by Abnar & Zuidema aclweb.org/anthology/2020.ac…
6
19
Here's an update from Dan Jurafsky and the #acl2020nlp team re COVID19: acl2020.org/ #NLProc
Screen cap of the announcement: "While we are looking into all possible options (including an all-virtual conference) we think by far the most likely scenario for ACL2020 is that we will have our regular live conference in July as usual. We do expect somewhat smaller attendance, ..."

ALT Screen cap of the announcement: "While we are looking into all possible options (including an all-virtual conference) we think by far the most likely scenario for ACL2020 is that we will have our regular live conference in July as usual. We do expect somewhat smaller attendance, ..."

2
36
4
110
Load more