R to @aisongcontest: And who's in the jury for the #AISongContest? A panel of AI bots who moonlight as music critics? 'Your song lacks emotional depth,' says Siri, probably. 🤖🎶

R to @dadabots: just seeing this, but lmk if you decide to try this and i can help at all! fwiw the annotation costs are probably higher than the actual training cost.

R to @aisongcontest: they all SUCK!

R to @dadabots: we're dope and awesome (which we are), which mean we haven't got enough time for doing all we want 😄

Mega mucho thanks to @dadabots @halving_fun and 1 anon for picking up editions of Ancient Waves Of Flowers! 🙏🙏💜💜💜🙏🙏 3 days left to get yours!

R to @dadabots: 🙂

If you’re dope and awesome (which you are), that means anything you do is dope and awesome 🤔

R to @LiuHaohe: >@Web3Metal

RT by @dadabots: I think I had a hand in this

R to @antoniohyder: yo @Garcio72 by any chance will you be attending ? let me treat you. pd. super f* masked of course

RT by @dadabots: 🎉 Enjoy the Top 10 of the 2023 #AISongContest #WalkingWithAI 🎵 Listen to all and vote your favourite until October 30 👉🏼 https://cutt.ly/9wWxgrui 🎟️ Get your tickets since 8 euros 👉🏼 https://cutt.ly/awWxfJ1h 🎤 Are you press? Apply for accreditation 👉🏼 https://cutt.ly/8wWxhFgJ

RT by @dadabots: "Sure, the comedown on MDMA is terrible, but have you ever been rate limited by DALLE-3?"

R to @matthewchaim: same. Unlike money, love is abundant and not zero sum. Maybe even love accelerates on itself like a singularity too.

RT by @dadabots: The following were invented by non-profit researchers: - Alibi - Scaled RoPE - Flash Attention - Parallel attention and MLP layers - Latent diffusion - CLIP-Guidance - Training jointly on code and NL - GPT-NeoX tokenizer - 4-bit quant and qLoRA - RWKV - Instruction-tuning

R to @dadabots: ☠️🤘🏽✊🏼

RT by @dadabots: 🎶Thrilled to share that my labmates at #MusicXLab have enriched #MusicGen with content-based controls with - chords🎵 - drums 🥁 - piano rolls🎹! Stay tuned for their upcoming paper and model release on GitHub & HuggingFace!🚀#MusicNLP #AudioGeneration https://kikyo-16.github.io/coco-mulla/

RT by @antoniohyder: Since 1913 @ArtinAmerica has been a monthly magazine about art, and here we are today drooling in its newsletter🙏 thank you Andy ❤️ https://view.email.artnews.com/?qs=7290f1f5e121968b26b7d1b730f5d072eea53faed9f0196e92e3e047c1ab526f6e48d5f9f1f6d688889230945f8808ce2675fa529c5a5e3fd872f4c3a4ba3b834de2f4fcbf9c9cbe

RT by @hafidhsoekma: We released Stable Audio Tools repo! Basically it's our training code for large latent diffusion audio models: - conditional/unconditional - autoencoders (vae, rvq, dac/encodec/oobleck) - json configs - gradio inference - model parallelism - no models yet https://github.com/Stability-AI/stable-audio-tools

RT by @filthy_priest: 🎉 Announcing http://stableaudio.com v1.0 🥳 - 95 second music generation - Latent diffusion model (like stable diffusion) - Can do genre fusions - Trained by @harmonai_org