Ben Hayes
@ben-hayes.bsky.social
Machine learning for audio synthesis @ Sony CSL Paris
PhD @ C4DM, QMUL.
Former intern at Spotify, Sony CSL, Bytedance
PhD @ C4DM, QMUL.
Former intern at Spotify, Sony CSL, Bytedance
Very excited to share that our latest work, "Audio synthesizer inversion in symmetric parameter spaces with approximately equivariant flow matching", has been accepted to ISMIR 2025 in Daejon, Korea!
Paper: arxiv.org/abs/2506.07199
Audio: benhayes.net/synth-perm/
Code: github.com/ben-hayes/sy...
🧵
Paper: arxiv.org/abs/2506.07199
Audio: benhayes.net/synth-perm/
Code: github.com/ben-hayes/sy...
🧵
June 10, 2025 at 10:13 AM
Very excited to share that our latest work, "Audio synthesizer inversion in symmetric parameter spaces with approximately equivariant flow matching", has been accepted to ISMIR 2025 in Daejon, Korea!
Paper: arxiv.org/abs/2506.07199
Audio: benhayes.net/synth-perm/
Code: github.com/ben-hayes/sy...
🧵
Paper: arxiv.org/abs/2506.07199
Audio: benhayes.net/synth-perm/
Code: github.com/ben-hayes/sy...
🧵
Reposted by Ben Hayes
DiffVox integrates differentiable vocal effects; analysis reveals parameter correlations and connections to McAdams' timbre dimensions; parameter distributions non-Gaussian; code and datasets available.
DiffVox: A Differentiable Model for Capturing and Analysing Professional Effects Distributions
Chin-Yun Yu, Marco A. Martínez-Ramírez, Junghyun Koo, Ben Hayes, Wei-Hsiang Liao, György Fazekas, Yuki Mitsufuji
arxiv.org
April 22, 2025 at 8:48 AM
DiffVox integrates differentiable vocal effects; analysis reveals parameter correlations and connections to McAdams' timbre dimensions; parameter distributions non-Gaussian; code and datasets available.
Generative modelling in latent space
Latent representations for generative models.
sander.ai
April 15, 2025 at 11:23 AM
amazing how the soothing beep of stolen Lime bikes has so naturally woven itself into the London soundscape
April 12, 2025 at 6:14 PM
amazing how the soothing beep of stolen Lime bikes has so naturally woven itself into the London soundscape
turned on an old computer and found some old unfinished music gathering dust. uploading it so it at least lives somewhere.
hard drive clear out 2016-2020, by Ben Hayes
21 track album
benhayes.bandcamp.com
April 6, 2025 at 4:11 PM
turned on an old computer and found some old unfinished music gathering dust. uploading it so it at least lives somewhere.
realised tonight there are only 3 red hot chili peppers songs:
1. california
2. zoop di blamp
3. heroin, but it's a woman
1. california
2. zoop di blamp
3. heroin, but it's a woman
March 29, 2025 at 12:23 AM
realised tonight there are only 3 red hot chili peppers songs:
1. california
2. zoop di blamp
3. heroin, but it's a woman
1. california
2. zoop di blamp
3. heroin, but it's a woman
Reposted by Ben Hayes
A low-latency neural audio synthesizer (BRAVE) was designed by analyzing latency sources in existing models (RAVE); BRAVE improved pitch and loudness replication while maintaining timbre modification capabilities, implemented in a specialized inference framework.
Designing Neural Synthesizers for Low Latency Interaction
Franco Caspe, Jordie Shier, Mark Sandler, Charalampos Saitis, Andrew McPherson
arxiv.org
March 17, 2025 at 11:08 AM
A low-latency neural audio synthesizer (BRAVE) was designed by analyzing latency sources in existing models (RAVE); BRAVE improved pitch and loudness replication while maintaining timbre modification capabilities, implemented in a specialized inference framework.
negative \vspace season approaches 😈
March 5, 2025 at 3:45 PM
negative \vspace season approaches 😈
Reposted by Ben Hayes
NablAFx, an open-source PyTorch framework, supports differentiable black-box and gray-box modeling of audio effects; it includes model architectures, datasets, training features, and plotting functions.
NablAFx: A Framework for Differentiable Black-box and Gray-box Modeling of Audio Effects
Marco Comunità, Christian J. Steinmetz, Joshua D. Reiss
arxiv.org
February 18, 2025 at 10:48 AM
NablAFx, an open-source PyTorch framework, supports differentiable black-box and gray-box modeling of audio effects; it includes model architectures, datasets, training features, and plotting functions.
Reposted by Ben Hayes
🎶✨ New Paper Announcement! ✨🎶
We present "Improving Musical Accompaniment Co-creation via Diffusion Transformers" 🎹🎸—a study advancing our Diff-A-Riff stem generator through improved quality, efficiency, and control.
📜Read the full paper here: arxiv.org/pdf/2410.23005 🧵👇
We present "Improving Musical Accompaniment Co-creation via Diffusion Transformers" 🎹🎸—a study advancing our Diff-A-Riff stem generator through improved quality, efficiency, and control.
📜Read the full paper here: arxiv.org/pdf/2410.23005 🧵👇
arxiv.org
January 20, 2025 at 1:42 PM
🎶✨ New Paper Announcement! ✨🎶
We present "Improving Musical Accompaniment Co-creation via Diffusion Transformers" 🎹🎸—a study advancing our Diff-A-Riff stem generator through improved quality, efficiency, and control.
📜Read the full paper here: arxiv.org/pdf/2410.23005 🧵👇
We present "Improving Musical Accompaniment Co-creation via Diffusion Transformers" 🎹🎸—a study advancing our Diff-A-Riff stem generator through improved quality, efficiency, and control.
📜Read the full paper here: arxiv.org/pdf/2410.23005 🧵👇
Reposted by Ben Hayes
speaking at Akademie der Bildenden Künste in Munich on Dec 16th
"Phantasmagoria: Sound Synthesis after the Turing Test"
about the methodological, ethical, and environmental implications of Generative AI for audio
by invitation from Florian Hecker
hal.science/hal-04650754
"Phantasmagoria: Sound Synthesis after the Turing Test"
about the methodological, ethical, and environmental implications of Generative AI for audio
by invitation from Florian Hecker
hal.science/hal-04650754
December 6, 2024 at 11:30 AM
speaking at Akademie der Bildenden Künste in Munich on Dec 16th
"Phantasmagoria: Sound Synthesis after the Turing Test"
about the methodological, ethical, and environmental implications of Generative AI for audio
by invitation from Florian Hecker
hal.science/hal-04650754
"Phantasmagoria: Sound Synthesis after the Turing Test"
about the methodological, ethical, and environmental implications of Generative AI for audio
by invitation from Florian Hecker
hal.science/hal-04650754
Reposted by Ben Hayes
DMRN+19 workshop registration has opened!
Keynote speakers: Stefan Lattner (Research Leader at Sony CSL Paris): Models of Musical Signals: Representation, Learning & Generation
www.qmul.ac.uk/dmrn/dmrn19/
Keynote speakers: Stefan Lattner (Research Leader at Sony CSL Paris): Models of Musical Signals: Representation, Learning & Generation
www.qmul.ac.uk/dmrn/dmrn19/
DMRN+19: Digital Music Research Network One-day Workshop 2024 - Digital Music Research NetworkFacebookTwitter XInstagramYouTubeLinkedInTikTok
www.qmul.ac.uk
December 3, 2024 at 4:47 PM
DMRN+19 workshop registration has opened!
Keynote speakers: Stefan Lattner (Research Leader at Sony CSL Paris): Models of Musical Signals: Representation, Learning & Generation
www.qmul.ac.uk/dmrn/dmrn19/
Keynote speakers: Stefan Lattner (Research Leader at Sony CSL Paris): Models of Musical Signals: Representation, Learning & Generation
www.qmul.ac.uk/dmrn/dmrn19/