Log inSign up
user avatar
Jesse Engel
@jesseengel
Guitarist, Researcher Google DeepMind. Opinions are my own. @jesseengel.bsky.social @[email protected]
Joined April 2009
65
Following
9,217
Followers
  • Pinned
    user avatar
    Jesse Engel
    @jesseengel
    Jun 4
    Had one of those "Oh, I'm living in the future" moments yesterday. Flying on the airplane. No internet. Playing music with hand gestures controlling a realtime LLM running locally on my laptop Person next to me thinking I'm crazy... all the while I'm having a great time 😅
    Image
    00:00
    23K
  • user avatar
    Jesse Engel
    @jesseengel
    Oct 12, 2018
    I've started mentally replacing "AI", "Neural Network", and other buzz terms with "Predictive model". It removes a false sense of agency, better reflects reality, and instantly brings the hype factor down by 300%.
  • user avatar
    Jesse Engel
    @jesseengel
    Jan 15, 2020
    Differentiable Digital Signal Processing (DDSP)! Fusing classic interpretable DSP with neural networks. ⌨️ Blog: magenta.tensorflow.org/ddsp 🎵 Examples: g.co/magenta/ddsp-e… ⏯ Colab: g.co/magenta/ddsp-d… 💻 Code: github.com/magenta/ddsp 📝 Paper: g.co/magenta/ddsp-p… 1/
    Image
  • user avatar
    Jesse Engel
    @jesseengel
    Feb 28, 2019
    Make music with GANs! GANSynth is a new method for fast generation of high-fidelity audio. 🎵 Examples: goo.gl/magenta/gansyn… ⏯ Colab: goo.gl/magenta/gansyn… 📝 Paper: goo.gl/magenta/gansyn… 💻 Code: goo.gl/magenta/gansyn… ⌨️ Blog: magenta.tensorflow.org/gansynth 1/
    Image
  • user avatar
    Jesse Engel
    @jesseengel
    Aug 7, 2025
    Realtime interactive generative models FTW! Announcing a new 🌊 of details and features for Magenta RealTime, the open weights live music AI model from GDM! * Live Jamming with audio input 🎤🎸🎵 * Personalize your own models 🔧 * Tech report 📜 Links below in the 🧵...
    Image
    00:00
    183K
  • user avatar
    Jesse Engel
    @jesseengel
    Sep 18, 2021
    It’s well known that neural networks model correlation and not causation. Recently, I’ve found it helpful to think about NN blocks as literal correlations of correlations of correlations … (incl. dense, norm, nonlin, conv, softmax, transformer, LMs, GANs, …) 🧵 1/20
  • user avatar
    Jesse Engel
    @jesseengel
    May 1, 2020
    A lot of folks have been asking me my thoughts about the recent Jukebox work by @OpenAI, so I thought a thread might help. I feel like I have separate reactions from three different parts of my identity: 1) ML researcher 2) ML researcher of music 3) Musician Long thread :) 1/17
  • user avatar
    Jesse Engel
    @jesseengel
    Jun 8, 2022
    DDSP-VST! Realtime Neural Audio Synthesis as an audio plugin. Excited to finally release this fun way to explore new sounds. 🔊 Plugin: g.co/magenta/ddsp-v… 🚂Train Models: g.co/magenta/train-… 🎮Discord: discord.gg/eyzhzMJMx5 📝Blog: magenta.tensorflow.org/ddsp-vst-blog 1/
    Image
    00:00
  • user avatar
    Jesse Engel
    @jesseengel
    Oct 11, 2020
    I'm not sure who needs to hear this, but just to be clear, music will never be "solved” because music is not a "problem". It is a solution. We'll always make music regardless of how good a model gets, because making and sharing music is an unfailing cure for the human condition.
  • user avatar
    Jesse Engel
    @jesseengel
    Aug 27, 2021
    🚨New Job Alert🚨 Passionate about the intersection of creativity🎵🎨, machine learning🤖, and human interaction👩‍🎨👨‍🎤? We're hiring a Research Scientist at Magenta! careers.google.com/jobs/results/1…
  • user avatar
    Jesse Engel
    @jesseengel
    Feb 25, 2019
    Modular reuse of generative models for cross-modal domain transfer (arxiv.org/abs/1902.08261). New work out from @alanyttian from his great Magenta internship. Without retraining the base models, you can get good domain transfer between very different domains (ex. pix2audio). 1/
    Image
    Image
  • user avatar
    Jesse Engel
    @jesseengel
    Oct 22, 2022
    No reason, just feel the need to say it again, there is no such thing as "an AI". It'd be so much clearer if we actually described what the tech is (Algorithms) and what it does (Automate). Automation Algorithms, or Augmentation/Assistance Algorithms depending on how we use them.
  • user avatar
    Jesse Engel
    @jesseengel
    May 26, 2022
    One takeaway for me from (#dalle2, #imagen, #flamingo) is there's no one "golden algorithm" to unlock these new transfer learning capabilities. Contrastive, AR, Freezing, Priors, they all can work. You almost can't stop these models from exhibiting these new types of behavior...
  • user avatar
    Jesse Engel
    @jesseengel
    Jan 12, 2025
    What an astonishingly terrible take. He misses the entire point of creative practice. Accessibility and mastery are not opposed, they're two ends of the same personal creative journey. Great tools help people start that journey (accessible), but also provide challenges that
    user avatar
    Michael Patti
    @mpatti
    Jan 11, 2025
    “It’s not really enjoyable to make music now… it takes a lot of time, it takes a lot of practice, you have to get really good at an instrument or really good at a piece of production software. I think the majority of people don’t enjoy the majority of time they spend making
    Image
    00:00
    17K