Paweł Cyrta

Researcher 👨‍🔬 (Speech 🗯️ & Music🎶 Technology) #TTS, #ASR #ML and 🎙️🎮🎚️🎛️🔊💬〰️📚📒📊📺📻

[Paweł Cyrta @cyrta]

Here is the link to slides 📔 from my talk🗣️ about
🎙️ Sound modeling 🎵 🔊 - latest techniques for music creation 🎛️. during @pydatawarsaw 2019.
It was great conference 🚀 and i am glad to meet 🤝 all those exciting people !

https://t.co/moAaRuaXtS

[Paweł Cyrta @cyrta]

@francoisfleuret There are a few attempts already on GitHub.
However, I would rather see it as a local app, that is fine tuning during the night locally, even if it lasts 2-4weeks people can wait, and it can be train on all local docs as well. Imho apple and Samsung are already doing such thing.

[Paweł Cyrta @cyrta]

@francoisfleuret They are already being made and new coming https://t.co/TUHjVs8eOW

[Tero Parviainen @teropa]

"We speak of 'playing' a violin, not 'working' it. If using a computer program feels like working in a bank or a hamburger chain restaurant, musicians won't (and shouldn't be asked to) do it."

Miller Puckette, Max at Seventeen https://t.co/aTzEBbKCri

[Paweł Cyrta @cyrta]

@hbredin Question is do we have as community a diverse dataset of decent size to compare, benchmark. They claim to have 250k. However 25k would be enough.

[Deepgram @DeepgramAI]

We put our and competitors’ diarization models through the test with over 250K human-annotated examples of real-word audio pulled of diverse audio domains, accents, speakers, and environments.

[Christian Steinmetz @csteinmetz1]

The Neural Audio Plugin Competition exceeded my expectations! I want to share all of the submissions with you in this thread. All the plugins are open source so you can check try them out for youself. It will be a long one...

[Paweł Cyrta @cyrta]

@chacinski Bardzo dziekuje za te miksy. Wyjątkowe to dla mnie są twoje odkrycia i wybory.
Btw. Mogę pomoc z automatycznym miksem całości jako jednej playlista albo jednego strumienia na YT. Dużo osób używa yt Music wraz z/bez premium. DM w razie czego.

[Paweł Cyrta @cyrta]

The tool to use is https://t.co/6FSTIhhPgA it looks great. I am fed up with hugginface frenzy

[Paweł Cyrta @cyrta]

That is great findings and initiative. Using distributed multitask training of many donors to work on model using flow like git. @colinraffel bravo https://t.co/LEABdb0JW6

[Yann LeCun @ylecun]

@rogue_node Oh, I'm a big fan of synthesizers, and I've been hacking synths since the 1970s.

[Tech Burrito @TechBurritoUno]

How resonance works

[Laura Gwilliams @GwilliamsL]

🧠how does the brain process rapid speech-sound sequences?

💬how is auditory content maintained over time?

🧮how is temporal order encoded using an evolving spatial pattern?

for answers, see our latest paper! https://t.co/IhirADnhFj @JeanRemiKing @davidpoeppel @AlecMarantz

[Sonicstate @sonicstate]

AUTOYT -sb22_b1_chair.mov ID- hCXmRZFm3-I (video): On a bike https://t.co/duw75ZcSzh

[Paweł Cyrta @cyrta]

@hbredin @ProjectJupyter Why wavesurfer, maybe directly in plotły or bohen

[Robert Scoble @Scobleizer]

The future of audio is Silicon speakers.

Yesterday I was invited to hear the future of audio on a visit to https://t.co/gzybKAcY7D

This Silicon wafer has hundreds of speakers on it. Yesterday they showed me these used in headphones, augmented reality glasses and more.

Wow.🧵

[Paweł Cyrta @cyrta]

@_BISHBash_ Could it be also stream or published afterwards ?

[Bartosz Ciechanowski @BCiechanowski]

Sound – https://t.co/PnbEJdYF9L

[Paweł Cyrta @cyrta]

Read my latest: “🕳️ Null, None, Nothingness ⬛ - Newsletter Pawel Cyrta 🔥 Wydanie #21 [2021W29] 💥” https://t.co/3wBOG7L26i

[Paweł Cyrta @cyrta]

Why Gen Z Doesn't Care About Music https://t.co/e0QhdDqR7F

[Paweł Cyrta @cyrta]

@csteinmetz1 @bustos_jor Yep, you can always create dynamic playlist like simple widget with only one audio object. It is still just a hack

[Paweł Cyrta @cyrta]

@maxrmorrison @bustos_jor The best i found is simple script that do import audio into html file as base64 and share it as file or url