Joined April 2008
- Tweets 1,677
- Following 603
- Followers 1,693
- Likes 1,272
147 Photos and videos
Jun 4
Music model you can actually play like a real instrument!
Running locally with basically zero latency is massive ๐Huge congrats to the @GoogleMagenta team! .๐ถ
Introducing Magenta RealTime 2 (MRT2): the live music model you can play as an instrument.
MRT2 offers MIDI and prompt controls, and runs natively on a MacBook with <200ms latency.
Open weights. Open source inference engine. Suite of apps and plugins.
Hear what it can do and try it out for yourself below ๐งต
1
11
551
Jun 4
SO cool! I Turning your digital life into a cartoon !!
Massive props to the DreamBeans team for such amazing work! ๐
Jun 3
๐จ NEW EXPERIMENT ๐จ
Dreambeans is a new, experimental mobile app that uses Personal Intelligence to connect to your Google apps. Every day, it delivers collections of personalized stories, surfacing things you might otherwise miss, alongside topics that are relevant to you, to help you dive deeper into the things you care about most.
Available starting today for eligible US-based Google AI Ultra users ( 18), with an open waitlist found on our website below!
Learn more at labs.google/dreambeans
1
2
313
Vivek Kumar retweeted
Apr 30
Not bad at allโฆ ๐๐
53
95
2,716
137,385
Vivek Kumar retweeted
Apr 13
This thing carries enough people to fill 10 full ride shares but it only provides one job. These eliminate 9 jobs each and the tech industry is trying to push them on my city
77
262
5,051
165,522
Vivek Kumar retweeted
ProducerAI is now Google Flow Music, and it supports editing as well as generation! See this post for details.
1/ A new era, a new name, and a new feature drop. ๐
ProducerAI is officially "Google Flow Music"! We're still the same team, with the same mission: building tools that push the boundaries of human creativity.
To kick off this new chapter, we're launching new capabilities to give you more control over your tracks.
Let's dive in: ๐งต๐
1
7
935
Apr 19
Google Flow Music - editing and generation both ๐๐๐
1/ A new era, a new name, and a new feature drop. ๐
ProducerAI is officially "Google Flow Music"! We're still the same team, with the same mission: building tools that push the boundaries of human creativity.
To kick off this new chapter, we're launching new capabilities to give you more control over your tracks.
Let's dive in: ๐งต๐
1
141
Vivek Kumar retweeted
Apr 14
Maybe tell your buddy to do some actual work and to stop spreading absolute nonsense. This post is completely false and just pure clickbait.
295
446
12,980
853,123
Mar 9
Congratulations @jesseengel and @GoogleDeepMind magenta team ๐
Opensourcing realtime lyria plugin is going to expand the artistic posibilities for musicians and creators looking to deepen their relationships with music using AI. Can't wait to see what people build ๐นโจ
Mar 9
Today, we're open sourcing the code behind "The Infinite Crate," our VST/AU plugin that lets you play with Lyria RealTime directly inside your favorite DAW! ๐ง๐ถ
Since we released The Infinite Crate last year, itโs been used by some of our favorite artists - including a wonderful showcase with @daitomanabe in Tokyo - and being featured as a top new music tool at NAMM 2026.
Weโre now fully open sourcing the plugin for developers to fork, modify, and make their own under the permissive Apache 2.0 license.
๐พ Get it here: github.com/magenta/the-infinโฆ
2
11
662
Feb 4
Our Sound Team at @GoogleDeepMind is hiring full-time Research Scientists in Mountain View & Cambridge, MA.
โ(listing below๐)
4
12
125
11,851
Vivek Kumar retweeted
Last night at #Sundance, we previewed Dear Upstairs Neighbors, an animated short film co-created by world-leading artists and our researchers.
This is true AI-empowered co-creation prioritising artist control: using custom AI workflows to turn hand-drawn storyboards into a full film. Read how we did it: blog.google/innovation-and-aโฆ
3
5
48
4,262
Jan 26
Blown away by what my colleagues have created here! ๐คฏ.
This level of innovation is exactly why I joined @GoogleDeepMind
Jan 26
Our short film Dear Upstairs Neighbors is previewing at @sundancefest. ๐ฌ
Itโs a story about noisy neighbors, but behind the scenes, itโs about solving a huge challenge in generative AI: control.
Developed by Pixar alumni, an Academy Award winner, researchers, and engineers, hereโs how it came together. ๐จ
6
528
2 Dec 2025
In San Diego this week for @NeurIPSConf Hit me up if you want to chat about anything Audio/GenAI๐๏ธ
Hard to believe it's been nearly a decade since my first one in 2016. The scale is completely different now. ๐คฏ #NeurIPS2025
1
7
597
Vivek Kumar retweeted
23 Oct 2025
I always thought the decline in fundamental AI research funding would happen because AI didnโt generate enough value to be worth the cost.
But it seems like itโs happening because it generated too much value. And the race to capture that value is taking priority.
Just remembering that a lot of this started in curiosity driven industry research labs.
32
36
363
71,344
10 Sep 2025
Excited to share our work from Sound Understanding team at @GoogleDeepMind!
Ever wanted to remove a single cough from a recording or make a faint doorbell louder? Recomposer makes editing complex audio scenes possible!
Paper: arxiv.org/abs/2509.05256
#AudioEditing #GenerativeAI
10 Sep 2025
๐New paper! Recomposer allows editing sound events within complex scenes based on textual descriptions and event roll representations. And we discuss the details that matter!
Work led by Dan Ellis w/ a bunch of Sound Understanding folks @GoogleDeepMind
arxiv.org/abs/2509.05256
1
5
25
3,742
10 Sep 2025
Recomposer uses "event roll" (a visual timeline of sounds) and simple text commands to guide edits. - giving precise control to delete, insert, or enhance specific sound events, like "enhance the Door sound from 1.5-2.0s."
An intuitive way to "recompose" a sound scene.
1
2
174
10 Sep 2025
Fantastic team effort. ๐
A huge thank you to the authors for their incredible work! ๐
Dan Ellis, @edfonseca_ @HakanErdoganPhD,
@DrJohnHershey, Ron J. Weiss, Kevin Wilson, @ScottTWisdom, Aren Jansen, @ChannningMoore & Manoj Plakal
@GoogleDeepMind
2
169
9 Sep 2025
AI for Finance done right ๐
Congrats @KrisBennatti and @HudsonLabs team ๐
9 Sep 2025
Most AI fails in finance โ
Hallucinations. Missed guidance. Inaccurate numbers.
Today we launch The Co-Analyst โ high-precision AI for institutional investors.
Used by funds managing $1T AUM.
Only the facts. Direct-from-source. Every time.
1
2
206
6 Aug 2025
Veo3-audio/Veo3-fast-audio at #1 ๐๐๐
๐ฌ The Video Arena Leaderboard is now live!
14,000 community votes have ranked the top Text-to-Video and Image-to-Video models.
๐ Text-to-Video rankings:
- #1 Veo3 (audio on)
- #3 Veo3, Veo3-fast
- #5 Hailuo 02 [Standard], Seedance 1.0 pro
- #6 Kling 2.1 Master
- #9 Wan 2.2 A14B
- #11 Pika 2.2, Mochi 1
Big congrats to @GoogleDeepMind, @Hailuo_AI, Bytedance, @Kling_ai, @Alibaba_Wan, @pika_labs, and @genmoai!
4
300
Vivek Kumar retweeted
24 Jul 2025
We just discovered the ๐ฅ COOLEST ๐ฅ trick in Flow that we have to share:
Instead of wordsmithing the perfect prompt, you can just... draw it. Take the image of your scene, doodle what you'd like on it (through any editing app), and then briefly describe what needs to happen (e.g. "changes happen instantly"). Using Frames to Video, Flow will understand the drawings and incorporate them into the final video.
Try it and show us what you find!
120
384
3,130
411,268