Musings about music production, psychoacoustics, sound synthesis & anything else interesting, from the Centre 4 Digital Music's Audio Engineering research team.
London, England
Joined July 2017
- Tweets855
- Following2.3K
- Followers1.5K
- Likes2.7K
QMUL_IntelSoundEng retweeted
🌟 Excited to Share Our Latest Work! 🎥🎶
Here we present Stable-V2A: Synthesis of Synchronized Sound Effects with Temporal and Semantic Controls
arxiv: arxiv.org/abs/2412.15023
web: ispamm.github.io/Stable-V2A
QMUL_IntelSoundEng retweeted
SyncFusion: Multimodal Onset-synchronized Video-To-Audio Foley Synthesis.
We use onsets extracted from silent videos and audio/text embeddings to condition a diffusion model for synchronized sound effects synthesis.
arxiv: arxiv.org/abs/2310.15247
web: mcomunita.github.io/diffusio…
QMUL_IntelSoundEng retweeted
This was work done during an internship @tapeitapp. A special thanks to my collaborator Thomas Walther and my advisor Josh Reiss (@IntelSoundEng). We will present this work at the upcoming AES Convention. Stop by if you are interested in controlling DSP with neural networks.
QMUL_IntelSoundEng retweeted
AI in music production: controversy & opportunity
ow.ly/G8uV50Pefve
@IntelSoundEng
QMUL_IntelSoundEng retweeted
Fuzz is not just a type of distortion effect, here we show why and in our #ICASSP2023 paper we capture its long-range input dependencies. Works for compression too.
paper: ieeexplore.ieee.org/document…
web: mcomunita.github.io/gcn-tfil…
vst: neutone.space/model/?id=9950…
QMUL_IntelSoundEng retweeted
Modulation Extraction for LFO-driven Audio Effects
Happy to share our work on extracting arbitrary LFO signals and modeling unseen audio effects!
arXiv: arxiv.org/abs/2305.13262
web: christhetr.ee/mod_extraction…
code: github.com/christhetree/mod_…
@csteinmetz1 @marcomunita @IntelSoundEng
QMUL_IntelSoundEng retweeted
Congratulations to 🇸🇪 🇫🇮 🇮🇱 🇮🇹 🇳🇴 🇪🇪 🇨🇿 , who I correctly predicted would finish in the top ten!
However, not everything about #Eurovision can be predicted. In the upcoming days, I'll be delving deeper into the results of my study.
#PhD @QMEECS @bbceurovision @Eurovision
QMUL_IntelSoundEng retweeted
Who will qualify to the Grand Final from the Eurovision 2nd Semi-Final tonight? 🎶I trained three models to predict which countries are the most likely to qualify. Will 🇨🇾 , 🇦🇲 , 🇵🇱 , 🇮🇸 , 🇦🇱 , 🇪🇪 , 🇬🇷 , 🇦🇹 and 🇷🇴 be going to the final? #Eurovision #EUROVISION2023
QMUL_IntelSoundEng retweeted
Funded PhD with me at @EdNapierSCEBE researching sound maps to make shared city spaces safer for #women and all citizens. tinyurl.com/3v4v4wjj
Please share!
Email [email protected] for more details.
#PhDposition #PhD #Equality #safety #sounddesign #empathy #education
QMUL_IntelSoundEng retweeted
DeepAFx-ST: Automatic control of audio effects for transferring the production style from one recording to another using differentiable signal processing.
arXiv: arxiv.org/abs/2207.08759
web: csteinmetz1.github.io/DeepAF…
GitHub: github.com/adobe-research/De…
@NicholasJBryan @IntelSoundEng
QMUL_IntelSoundEng retweeted
I am excited to speak about our work DeepAFx-ST at CCRMA at Stanford as part of the DSP Seminar series tomorrow at 3:30 PST (virtually).
The event is open to all and you can find the Zoom link at the page below.
ccrma.stanford.edu/events/de…
QMUL_IntelSoundEng retweeted
So honored to have been awarded best student paper at @AESorg NYC 2022! My paper with @marcomunita and @IntelSoundEng "Reverse Engineering Memoryless Distortion Effects with Differentiable Waveshapers" proposes parameterized nonlinearities in Wiener-Hammerstrien models. 🧵soon!
QMUL_IntelSoundEng retweeted
@josephtcolonel today presented our paper on “Reverse Engineering Memoryless Distortion with Differentiable Waveshapers” that won Best Paper Award at the @AESorg New York 2022 Convention
@CDT_AI_Music @c4dm @IntelSoundEng #AESshow #AESorg
QMUL_IntelSoundEng retweeted
Here we are, today is the day. Please check out the link and let us know what you think! producthunt.com/posts/roex-a…
Super excited to announce this Thursday we're planning to launch @roexaudio on @ProductHunt.
As you're probably aware, we've been building RoEx Automix: AI Powered Mixing Services for Musicians, Producers and Content Creators
Let us know what you think:
producthunt.com/posts/roex-a…
QMUL_IntelSoundEng retweeted
Happy to share our preprint on a novel #algorithm for #sound event detection that generalizes better and is faster. It is now a #regression task instead of frame-based #classification.
@djmoffat @bio_computer
#AudioSegmentation #SoundEventDetection
``You Only Hear Once: A YOLO-like Algorithm for Audio Segmentation and Sound Event Detection. (arXiv:2109.00962v1 [eess.AS]),'' Satvik Venkatesh, David Moffat, Eduardo Reck Miranda, ift.tt/3n5KFjs
QMUL_IntelSoundEng retweeted
There is a wonderful new paper on using natural language words to describe audio effects "Word Embeddings for Automatic Equalization in Audio Mixing", by @SatvikVenkatesh published in @AESorg Journal aes.org/e-lib/browse.cfm?eli… #audio #aes #research #mixing #semanticaudio #afx #PhD
QMUL_IntelSoundEng retweeted
``Automatic music mixing with deep learning and out-of-domain data. (arXiv:2208.11428v1 [eess.AS]),'' Marco A. Martínez-Ramírez, Wei-Hsiang Liao, Giorgio Fabbro, Stefan Uhlich, Chihiro Nagashima, Yuki Mitsufuji, ift.tt/iDeGQ5L
QMUL_IntelSoundEng retweeted
We're excited to announce that today we are soft-launching our initial offering of RoEx Automix!
For a short period of time, we'll also be giving away 1 mix credit with every sign-up. Sign up here automix.roexaudio.com/regist…
QMUL_IntelSoundEng retweeted
New article on Music Production Semantic and how language is used to describe different audio effect - The largest study ever conducted! aes.org/e-lib/browse.cfm?eli… @IntelSoundEng @BrechtDeMan @AESorg #Audio #musicproduction #semantics #audioeffects #research #openaccess