Francesc Lluis
@francesclluis_Deep learning for audio signal processing and acoustics @BangOlufsen.
Similar User
@edfonseca_
@carlthome
@jordiponsdotme
@di_bogdanov
@pablo__alonso
@porcaro_lorenzo
@guillemcs_
@felixml3
@nkundiushuti
@plachouras
@juan_s_gomez
@minje_research
@matangover
@pa9501460
@santty128
``Blind Spatial Impulse Response Generation from Separate Room- and Scene-Specific Information,'' Francesc Llu\'is, Nils Meyer-Kahlen, ift.tt/iwrvtHB
``Vocos: Closing the gap between time-domain and Fourier-based neural vocoders for high-quality audio synthesis,'' Hubert Siuzdak, ift.tt/YVzRsSG
1/7 For the past decade, our team at Meta Reality Labs (previously CTRL-labs) has been dedicated to developing a neuromotor interface. Our goal is to address the Human Computer Interaction challenge of providing effortless, intuitive, and efficient input to computers.
New (2h13m 😅) lecture: "Let's build the GPT Tokenizer" Tokenizers are a completely separate stage of the LLM pipeline: they have their own training set, training algorithm (Byte Pair Encoding), and after training implement two functions: encode() from strings to tokens, and…
well diffusion transformer was rejected at CVPR 2023 due to limited novelty.
R2: While the results are impressive, this is a simple combination of diffusion transformer (ICCV 2023) and latent diffusion model (CVPR 2022). Limited novelty. Weak reject.
I completely agree once you fully understand the true functions of the building (transformer) blocks in these models: They learn the distribution of the data (visual or text) given to them via compression (denoising for one), and then resample from the learned distribution.
LLM OS. Bear with me I'm still cooking. Specs: - LLM: OpenAI GPT-4 Turbo 256 core (batch size) processor @ 20Hz (tok/s) - RAM: 128Ktok - Filesystem: Ada002
This one's easy! That honour goes to "the diffusion bible", as I like to call it. It's been well over a year and I still refer to it several times a week. Very few papers I've read come close, in terms of signal-to-noise ratio. arxiv.org/abs/2206.00364
what paper (not your own, maybe not even in your own area) can you not stop telling people about?
SCOOP: this is BIG. Bellman Conformal Inference: Calibrating Prediction Intervals For Time Series” 🔥🔥🔥🔥🔥🚀🚀🚀🚀🚀. @Stanford paper by @ZitongYang0, @lihua_lei_stat and the great Emmanuel Candes. 🚀 Bellman Conformal Inference (BCI), a framework that wraps around any time…
6 PhD positions open in the context of the UPF-BMAT Chair on AI and Music, starting in October 2024. Applications until March 19th: upf.edu/web/mtg/home/-…
A rare opportunity for an industrial #PhD in #AI and #acoustics with @BangOlufsen is now open for applications 🚀 career5.successfactors.eu/sfcareer/jobre…
magnet:?xt=urn:btih:5546272da9065eddeb6fcd7ffddeef5b75be79a7&dn=mixtral-8x7b-32kseqlen&tr=udp%3A%2F%2Fopentracker.i2p.rocks%3A6969%2Fannounce&tr=http%3A%2F%https://t.co/g0m9cEUz0T%3A80%2Fannounce RELEASE a6bbd9affe0c2725c1b7410d66833e24
The people mocking "no proof" didn't realize that it was their insistence to have a particular type of theory that was holding them back. They would only work on "trivial" models for which the loss was convex, so they could use fancy semi-definite programming methods and prove…
98 years ago today, our founders Peter Bang and Svend Olufsen created Bang & Olufsen. Their first task: creating a radio that could run on batteries. Celebrate our birthday with us, and watch the full story here: youtu.be/C2w2-Un36pY #BangOlufsenLegacy #BangOlufsen
Interactive chat mode added to 🦙.cpp It actually works surprisingly well from the few tests that I tried! Kindly contributed by GH user Blackhole89
When you buy a record or CD, you own it, thanks to copyright's "first sale" principle. I have criticisms of copyright law, but at least it's created by a democratically accountable legislature. When you buy a digital download, your use is governed by private ToS, not law. 1/
Apart from noise, speech audio can feature reverb, clipping, narrow bandwidth, codec artifacts, etc. Here is UNIVERSE to rule them all! w/ @santty128, @jordiponsdotme, @r_oguz_araz, & D. Scaini. Paper: arxiv.org/abs/2206.03065 Examples: serrjoa.github.io/projects/unive… 🧵 1/5
I am looking for an audio-focussed #software guru to work on Next Gen #spatial #Audio, in a super cool team. Preferably skilled in C/C++ but not required #job #sound #acoustics - Drop a DM if you are the one!
Alert, musicologists! Exciting opportunity to work with us on a truly interdisciplinary digital musicology (+ music informatics, performance science, cultural heritage, digital scholarly dissemination, semantic publishing) project! Plus, Vienna is a very lovely place to live 🇦🇹🎡
🚨 3-year funded PhD opportunity: digital musicology, performance science, @MusicEncoding Join us at @mdwwien to analyse the signature sound of the @Vienna_Phil New Year's Concert series. Is it "the same procedure (as) every year?" tinyurl.com/signaturesound Please RT! 🎻🥂🎶
United States Trends
- 1. Jake Paul 1,03 Mn posts
- 2. #Arcane 205 B posts
- 3. Jayce 41,7 B posts
- 4. Serrano 246 B posts
- 5. #SaturdayVibes 2.490 posts
- 6. Vander 13,4 B posts
- 7. Good Saturday 22,2 B posts
- 8. #HappySpecialStage 79,3 B posts
- 9. maddie 17,9 B posts
- 10. #saturdaymorning 1.798 posts
- 11. Jinx 98,5 B posts
- 12. Isha 31,5 B posts
- 13. #SaturdayMood 1.182 posts
- 14. Canelo 17,5 B posts
- 15. Rizwan 7.585 posts
- 16. Father Time 10,7 B posts
- 17. Woop Woop 1.170 posts
- 18. The Astronaut 28,6 B posts
- 19. Super Tuna 21 B posts
- 20. Babar 11 B posts
Who to follow
-
Eduardo Fonseca
@edfonseca_ -
Carl Thomé
@carlthome -
Jordi Pons
@jordiponsdotme -
Dmitry Bogdanov
@di_bogdanov -
Pablo Alonso Jiménez
@pablo__alonso -
Lorenzo Porcaro
@porcaro_lorenzo -
Guillem Cortès Sebastià
@guillemcs_ -
felixml
@felixml3 -
Marius Miron
@nkundiushuti -
Christos Plachouras
@plachouras -
Juan Sebastián Gómez-Cañón
@juan_s_gomez -
Minje Kim
@minje_research -
Matan Gover
@matangover -
Arijit Biswas
@pa9501460 -
Santi PdP
@santty128
Something went wrong.
Something went wrong.