Tero Parviainen's banner

Tero Parviainen

@teropa • 9,829 subscribers

One half of @ctpt_studio. Building music systems for wellbeing and movement.

Videos

Anya Rossi

sweetdream.ai

SweetDream.ai•Sponsored•Livecam

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

Exclusive private shows

1.2k viewers online

Private Show

Join now for exclusive access

Free preview available • Premium content

Terence Tao says evolution hasn't given us a "math center" in the brain, so different people repurpose different areas of the brain to solve math problems. Which leads to different styles of learning and communication.

Terence Tao says evolution hasn't given us a "math center" in the brain, so different people repurpose different areas of the brain to solve math problems. Which leads to different styles of learning and communication.

Tero Parviainen

132,896 views • 3 months ago

you know where in-the-moment, "system 1" type ai decision-making could be cool? music technology. jevseq is a demented multiplayer drum sequencer. you paint the world, jev walks it in real time, and tries its hardest to hit a snare on 2 and 4.

you know where in-the-moment, "system 1" type ai decision-making could be cool? music technology. jevseq is a demented multiplayer drum sequencer. you paint the world, jev walks it in real time, and tries its hardest to hit a snare on 2 and 4.

Tero Parviainen

34,526 views • 1 month ago

what i do while claude is working

what i do while claude is working

Tero Parviainen

218,148 views • 8 months ago

Going to use this next time I hear the "machines learn from looking at images just like human artists do, ergo it's fine for corpos to just take all creative work they find on the internet" type of argument

Going to use this next time I hear the "machines learn from looking at images just like human artists do, ergo it's fine for corpos to just take all creative work they find on the internet" type of argument

Tero Parviainen

30,737 views • 3 months ago

in which Jeremy Howard nails the phenomenology of agentic coding

in which Jeremy Howard nails the phenomenology of agentic coding

Tero Parviainen

41,792 views • 6 months ago

Have been doing some stem remixing work with Stable Audio 3. 📦 Medium model 🔉 init_audio holding the original audio file 😶‍🌫️ init_noise_level between 0.4-0.5 seems to be the sweet spot 🪄 Empty promps

Have been doing some stem remixing work with Stable Audio 3. 📦 Medium model 🔉 init_audio holding the original audio file 😶‍🌫️ init_noise_level between 0.4-0.5 seems to be the sweet spot 🪄 Empty promps

Tero Parviainen

26,716 views • 4 months ago

TIL the Finnish public service broadcaster employs musicians during cross-country skiing competitions to perform the sounds of skiing in real time. They do this for hours at a time and have been doing it for decades.

TIL the Finnish public service broadcaster employs musicians during cross-country skiing competitions to perform the sounds of skiing in real time. They do this for hours at a time and have been doing it for decades.

Tero Parviainen

76,617 views • 2 years ago

I which Sander Dieleman beautifully illustrates why diffusion models work so well with images 🐳 Our visual world is spatially coherent, and large scale structures dominate. 〜 This means low frequency spectral components in images tend to be stronger. 🌫️ This means when you noise up an image, high frequency detail goes away first, then coarse structure. 😶‍🌫️ This means when you denoise an image, the coarse structure emerges first, then the fine detail. ♻️ This means you can think of diffusion as "spectral autoregression". Coarse / low end to fine / high end.

I which Sander Dieleman beautifully illustrates why diffusion models work so well with images 🐳 Our visual world is spatially coherent, and large scale structures dominate. 〜 This means low frequency spectral components in images tend to be stronger. 🌫️ This means when you noise up an image, high frequency detail goes away first, then coarse structure. 😶‍🌫️ This means when you denoise an image, the coarse structure emerges first, then the fine detail. ♻️ This means you can think of diffusion as "spectral autoregression". Coarse / low end to fine / high end.

Tero Parviainen

16,450 views • 4 months ago

Using Stable Audio 3 to generate variations of an existing loop. Unconditional generation (no prompt), renoising the latents to 0.5, and just using different seeds seems to generate a nice neighbourhood around the original. Generally keeps the harmonic context and feel.

Using Stable Audio 3 to generate variations of an existing loop. Unconditional generation (no prompt), renoising the latents to 0.5, and just using different seeds seems to generate a nice neighbourhood around the original. Generally keeps the harmonic context and feel.

Tero Parviainen

10,894 views • 5 months ago

your regular reminder on just how fucking FUN is

your regular reminder on just how fucking FUN is

Tero Parviainen

21,039 views • 3 years ago

"Making Minimal Techno with WebAssembly And a C Compiler" My presentation from #JSHeroes2024 today 👉

"Making Minimal Techno with WebAssembly And a C Compiler" My presentation from #JSHeroes2024 today 👉

Tero Parviainen

15,603 views • 2 years ago

No more content to load