David Marttila

apropos of nothing:

this is an appreciation post for @simonwillison.net, who tries out literally all the tools, and joyfully shares (for free!) what he learns in the hope that it will be useful to other humans

grateful to have someone relentlessly building with the garage door open 💗🙌

A friend got me onto espresso tonic. Such a perfect summer drink.

It's also pretty reductive though to just look at the "boots on the moon" result of spacefaring technology. GPS alone is a massive one.

Dass Waymo seit 5 Jahren autonome Taxis ohne Sicherheitsfahrer operiert ist egal?

A, I guess, if the distinction to B is lack of a human-in-the-loop

Something genuinely cool can come out of a human working with generative AI, like those QR codes from 2 years ago

Kyiv, Odesa: Russian terrorists targeted residential buildings overnight. Entire apartment sections have collapsed in some areas, the number of casualties has surpassed one hundred.

The whole thread is worth a read but using Hobbes as an analogy for human/AI assistant relations is lowkey brilliant

Super interesting! Would you happen to know if similar work has been done for auditory perception?

(Whispering quietly)

Ready Player One

Datadog released an open-weight, state-of-the-art time series foundation model called Toto, designed for multivariate time series forecasting with an emphasis on observability metrics.

They also introduced BOOM, their time series observability benchmark.

I've been really enjoying working with JAX recently and I'm happy that I could make a tiny contribution to improve audio workflows; from the next release on, expect about a 10x speedup in jax.scipy.signal.stft!

github.com/jax-ml/jax/p...

Holy crap! The vocal effects plugin I wrote for BABY audio as a collaboration during my Master's at AAU Copenhagen is currently #4 in the iPad app store charts for paid music apps 🤯

Complex www.smbc-comics.com/comic/ivy-le...

It's hard to overstate how dramatically pixi.sh and github.com/astral-sh/uv improve Python environment management. I've nearly entirely eliminated conda/mamba from my toolkit - and on the rare occasion I need one, I can trivially generate an environment from pixi.

Black-box and gray-box models for nonlinear audio effects were compared on a large dataset (ToneTwist AFx); time-varying gray-box models were introduced, and subjective evaluation was conducted.

YouTube video by C4DM - Centre for Digital Music Models of Musical Signals: Representation, Learning & Generation. Stefan Lattner (Sony SCL). DMRN+19

Video of @stefanlattner.bsky.social 's talk at DMRN+19 is finally online:
"Models of Musical Signals: Representation, Learning & Generation" @c4dm.bsky.social
www.youtube.com/watch?v=ixHf...

Writing a Python script that outputs CMakeLists.txt - in other words, I'm writing code to output some code that will output some code that will compile my code. Modern computing is weird.

Anabel and I rewatch Buffy regularly, and the thing that always impresses me is how regardless of varying strength of budget, writing, other actors, etc, Sarah Michelle Gellar gives it her all in every single scene, every second she's onscreen. Funny, sincere, heroic. Cleans her plate.

This feels like it was designed in a lab specifically to appeal to every possible facet of my musical taste.

When I reduce the amount to below the "free shipping" threshold, I can actually open PayPal. Could you get someone to look into that? Thanks!!

Hi @hankgreen.bsky.social! I'm seeing an error at checkout when ordering to the UK. all three payment methods just fail silently and don't let me proceed. I suspect it might be related to the delivery address, since the cart applies the free shipping discount even though it's not a domestic order.

For anyone interested in #audioprogramming, #audiotech, #musicai, #musictech, #aimusic research: @arxiv-sound.bsky.social is now live. Inspired by, but not related to, the similar account on the former bird app.

This probably won't fit the list but might still be of interest to some people reading this: I just made an account that automatically posts arXiv articles related to sound, music and speech signal processing to bluesky: @arxiv-sound.bsky.social