Roger Tseng
35 posts
Roger Tseng
@RogerTseng
Speech Processing & Machine Learning Lab @ntu_spml | National Taiwan University
Katılım Ağustos 2009
86 Takip Edilen29 Takipçiler
@Musical_sophie Hello! May I ask if it's still available? Sorry for the public comment, can't message you
English
@itsaidanneee_ Yes! Could you send me a message so we can work out the details?
English
Code, Video & Poster: github.com/roger-tseng/sp…
Full paper: arxiv.org/abs/2303.08809
If you find this topic interesting, come chat!
I'll present our paper tomorrow from 3:40 PM - 5:10 PM at the Audio and Text Segmentation, Tagging and Parsing (SLT-P9) poster session at #ICASSP2023
English
Another interesting discovery I wish to highlight is that Approach 2, when trained without any textual data, still shows signs of learning the different branching directions (a lang.-specific syntactic property) of English and Korean from speech directly.
English
Unsupervised models that learn from raw speech have been limited to discovering acoustic/lexical information, such as phonemes or words.
Can we build unsupervised speech models that learn high-level linguistic information like syntax?
#ICASSP2023
English
Roger Tseng retweetledi
The research topic I've been working on for a long time is going to be incorporated into iOS 17 by Apple (of course, I didn't write the code in iOS): 9to5mac.com/2023/05/16/app…
English
Roger Tseng retweetledi
``Analysing Discrete Self Supervised Speech Representation for Spoken Language Modeling. (arXiv:2301.00591v1 [cs.CL]),'' Amitay Sicherman, Yossi Adi, ift.tt/KwI4mGs
English
Roger Tseng retweetledi
Abdelrahman Mohamed (Meta), Shinji Watanabe (CMU), Tara Sainath (Google), Karen Livescu (TTIC), Shang-Wen Li (Meta), Shu-wen Yang (NTU), Katrin Kirchhoff (Amazon), and I will give a tutorial about self-supervised learning for speech at NAACL 2022. #t2" target="_blank" rel="nofollow noopener">2022.naacl.org/program/tutori…
English
Note to self to check this out later. What does this suggest for unsupervised language acquisition?
Sepp Hochreiter@HochreiterSepp
ArXiv arxiv.org/abs/2207.02098: Empirical study on NNs for learning formal languages. RNNs and Transformers fail to generalize on non-regular tasks, LSTMs solve regular and counter-language tasks, and only NNs with structured memory learn context-free and context-sensitive tasks.
English
Roger Tseng retweetledi
"Progressive Stage-wise Learning for Unsupervised Feature Representation Enhancement"
Someone finally beat regular backpropagation on a realistic problem—specifically, in the context of self-supervised learning (SSL). [1/12]
English
Note to self to check this out later. What does this mean for unsupervised language acquisition?
Sepp Hochreiter@HochreiterSepp
ArXiv arxiv.org/abs/2207.02098: Empirical study on NNs for learning formal languages. RNNs and Transformers fail to generalize on non-regular tasks, LSTMs solve regular and counter-language tasks, and only NNs with structured memory learn context-free and context-sensitive tasks.
English
Roger Tseng retweetledi
Great advice by @EugeneVinitsky
-- Run the simplest experiment first
-- Write a mini-version of your paper before getting started
-- Keep an ultra-detailed notebook
-- Every paper should be reproducible via 1 script
-- Save all your videos in one place
-- Present regularly
Eugene Vinitsky 🦋@EugeneVinitsky
A lot of blog posts on research vision, less so on the day to day of managing your research, experiments, and productivity. I’ve repeated this to students enough times so I’ve written it up tinyurl.com/Research-produ…
English
Roger Tseng retweetledi
``Self-supervised Learning with Random-projection Quantizer for Speech Recognition. (arXiv:2202.01855v2 [cs.CL] UPDATED),'' Chung-Cheng Chiu, James Qin, Yu Zhang, Jiahui Yu, Yonghui Wu, ift.tt/N1kKMaI
English
Roger Tseng retweetledi
A very interesting read! There's indeed a huge gap between usable tech a musician would enjoy and "fancy" tech the AI community finds worth investigating!🧐
On the Development and Practice of AI Technology for Contemporary Popular Music Production: transactions.ismir.net/article/10.533…
English