.@stableaudio curious about your product for hip hop prod. can you link me to tutorials on how to prompt the best? not having luck.

ついにabletonとstableaudioが接続できた♫

StableAudioどう使おうかな…と思ってたけど、ポモドーロ作業配信すればいいのか！休憩時間5分だけ話すのであればAPI代金も節約して長時間配信できるのいいかもな。

shoutout to @StabilityAI @stableaudio for open-sourcing stable audio 3 - clean model, clean representations, very fun to break open 🙏

【ComfyUIで音楽生成！Stable Audio最適化モデルが登場】 ComfyUI公式（Comfy-Org）によりStable Audio Openが再パッケージ化されました！ ファイル形式や内部構造が最適化されたことで、セットアップの難易度が大幅に下がっています。 プロンプトから高品質な音楽や効果音を生成できる強力な拡散モデル。 画像生成と同じ感覚で、コンテンツに最適な音を一瞬で作り出せます。 SNS動画のオリジナルBGM制作や、ゲーム開発用のSE量産に最適。 著作権フリーの高品質音源を自作して、クリエイティブの幅を広げましょう！🚀 #ComfyUI #StableAudio

【音楽生成の歴史が変わる！最新モデルStable Audio 3 Medium】 Stability AIから待望の実戦仕様モデルが登場しました！✨ 🔹 高速・高品質な生成 「潜在拡散モデル」により、ハイクオリティな音声を驚くほど短時間で作成できます。 🔹 自由な長さで編集可能 数秒のSEから長尺BGMまで対応。しかも既存の音素材をブラッシュアップする編集機能も備えています。 🔹 実践的な活用シーン ・YouTubeの完全オリジナルBGM制作 ・ゲーム開発における効果音の量産 ・広告動画に合わせた尺調整済みBGM販売 時間とコストを劇的に削減し、クリエイティブを次のステージへ引き上げる強力なツールです！🚀 #StableAudio #AI制作

ComfyUIによる音楽生成モデル「Stable Audio 3.0」のサポート開始。 ライセンス済み学習データによる商用利用への対応。 SmallモデルはCPUで最長約2分、Mediumモデルは最長約6分の効果音や楽曲生成。 #ComfyUI #StableAudio URLはリプ⬇️

Sony presents Woosh- sound effect foundation model. Something lika MMAudio. Generates T2A/V2A. - distilled variants hit 4 NFEs for instant inference; - looks better than StableAudio/TangoFlux. - 48kHz encoder/decoder, 3.7x compression; no code yet. sonyresearch.github.io/Woosh…

ゲームUPしたからそろそろ感想 一番頭を悩ませたのは効果音 …いや効果音作れるAI自体はあるのだがそれらで商用Freeが有料のStableAudioぐらいしかない ここだけは諦めてロイヤリティフリーの音源買って対処と なかなか全ては揃わないものだ

Ace-Step 1.5 just dropped and it’s a game-changer for ComfyUI. ✅ 2-minute high-quality songs in 22 seconds. ✅ Runs on a potato (4GB VRAM / RTX 3060). ✅ Features: Style LoRAs, Covers, and "Repainting" (fix lyrics!). It's fast, local, and free. Is this the end of Suno? 🎹 Full tutorial & workflow here: youtu.be/bVg3HZ4uPsg #ComfyUI #AceStep #AIMusic #GenerativeAI #Suno #StableAudio

TMG Sound as a Slow Variable Within Experience Music moves at its own pace ✅ Provide stability within experience ✅ Recurs gently across phases ✅ Extend naturally with time #SlowSound #SoundContinuum #StableAudio #ListeningFlow #MusicAsBuffer #UnhurriedExperience #SoundAndTime

Nice work from @fundwotsai2001 for adding time-varying controls to @stableaudio. ```This repo collects methods for following time-varying controls on Stable-Audio Open. We provide checkpoints on huggingface that follow melody condition (proposed in stable-audio-control).```

An example of AI vids locally in 20 minutes with open source models. The process: 1: Flux Krea model for start images. 2: Wan 2.2 I2V for video scenes. 3: StableAudio model for music 4: VibeVoice for narration of the video #ComfyUI #alibaba #huggingface #stabilityai #aivid

Arcade.Studio is AI world building done right! 🔥 Workflow bellow 👇 Only Arcade.Studio gives you access to all the AI modalities at once. Generate images, videos, music, and sound effects all in a refined workspace built for creative focus! Workflow ✨ ✨ ✨ 👉 To create the opening shot Click on the "Video" tab on the Arcade action bar and select Veo3 as your video model. Then use the following prompt: “A filthy, cluttered living room with a sagging recliner at center frame. The camera is static. A heavy man in a stained tank top, robe, and boxers shuffles in, slumps into the chair, lifts the footrest, and turns on the TV.” 👉 Then for the reverse angle of the TV with sock puppets, again we will choose Veo in the Arcade action bar and now use this prompt: “A grimy, dim room with an old CRT TV on a dirty floor amid trash and papers. The camera faces the screen showing a bright sock-puppet show, its flickering light illuminating the filthy walls.” 👉 For the next channel with the wrestler we will again use Veo with the prompt: “Close-up shot of a Mexican lucha libre wrestler inside a wrestling ring, wearing a colorful mask yelling into the camera.” 👉 For the next channel with the hotline woman asking you to call her, we first create the image of the woman. Click on the "Image" tab in the Arcade Action bar and choose your favorite image model. (nano banana, Seedance, or Imagen) Then use the following prompt: “Close-up of a blonde woman in a black dress under soft studio light, staring into the camera. 80s–90s VHS aesthetic with video noise, warm tones, blue backdrop, bold makeup, red lips — like a late-night hotline ad.” 👉 I then decided she should be brunet. To do this load the image in Nano Banana, Seedance, or Qwen and simple prompt: "Change her hair to brunet" 👉 Once we have the image of the woman we will create her voice in by click on the "Audio" tab in the Arcade Action bar. Just prompt what ever you'd like your actor to say! 👉 Now back to the "Video" tab and select OmniHuman, drag and drop the image of the woman and the voice you created to create a video of her speaking the dialog you created. 👉 For the final channel with the Alien new anchors we again choose Veo in the "Video" tab and use the following prompt: “Two alien news anchors at a retro 70s desk. First says, ‘This just in — Arcade Studio is the number one site for world building.’ Second replies, ‘Yes, even if that world is a gross dude’s apartment.’ Funny, cinematic sci-fi broadcast.” 👉 Editing was done in DaVinche DaVinci Resolve Give it a try!! 😁 arcade.studio 👉 Follow for more AI insights ✨ 👉 DM me for tips and tricks 😄 👉 Share with your team! 😎 #ArtificialIntelligence #AI #GenerativeAI #Animation #VideoProduction #Filmmaking #CreativeProcess #Design #Storytelling #ContentCreation #aivideo #vfx #motiondesign #worldbuilding #nanobanana #seedance #imagen #qwen #seedance #omnihuman #veo3 #kling #elevenlabs #stableaudio #arcadestudio

𝐒𝐭𝐚𝐛𝐥𝐞 𝐀𝐮𝐝𝐢𝐨 2.5: 𝐓𝐡𝐞 𝐧𝐞𝐰 𝐟𝐫𝐨𝐧𝐭𝐢𝐞𝐫 𝐢𝐧 𝐞𝐧𝐭𝐞𝐫𝐩𝐫𝐢𝐬𝐞-𝐠𝐫𝐚𝐝𝐞 𝐚𝐮𝐝𝐢𝐨 𝐠𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐨𝐧 Stability AI has introduced Stable Audio 2.5, which is designed specifically for enterprises that want high-quality, customizable audio fast. It helps create professional, commercial-grade sound that fits your brand everywhere, from ads to in-store experiences to digital content. Whether you need catchy music, immersive soundscapes, or a unique product sound, this tool can handle it. -- 𝐇𝐞𝐫𝐞’𝐬 𝐡𝐨𝐰 𝐢𝐭 𝐭𝐫𝐚𝐧𝐬𝐟𝐨𝐫𝐦𝐬 𝐞𝐧𝐭𝐞𝐫𝐩𝐫𝐢𝐬𝐞 𝐚𝐮𝐝𝐢𝐨 𝐩𝐫𝐨𝐝𝐮𝐜𝐭𝐢𝐨𝐧: 1. 𝐁𝐥𝐚𝐳𝐢𝐧𝐠 𝐅𝐚𝐬𝐭 𝐆𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐨𝐧 - Make audio tracks up to three minutes long in under two seconds using the ARC method. 2. 𝐒𝐨𝐩𝐡𝐢𝐬𝐭𝐢𝐜𝐚𝐭𝐞𝐝 𝐌𝐮𝐬𝐢𝐜𝐚𝐥 𝐂𝐨𝐦𝐩𝐨𝐬𝐢𝐭𝐢𝐨𝐧 - Generate full musical pieces with intros, developments, and outros based on your mood or style. 3. 𝐀𝐮𝐝𝐢𝐨 𝐈𝐧𝐩𝐚𝐢𝐧𝐭𝐢𝐧𝐠 𝐟𝐨𝐫 𝐂𝐫𝐞𝐚𝐭𝐢𝐯𝐞 𝐂𝐨𝐧𝐭𝐫𝐨𝐥 - Upload clips and tweak or expand them with intelligent audio inpainting. 4. 𝐂𝐮𝐬𝐭𝐨𝐦 𝐁𝐫𝐚𝐧𝐝 𝐒𝐨𝐧𝐢𝐜 𝐈𝐝𝐞𝐧𝐭𝐢𝐭𝐲 - Customize the model using your brand’s sounds to create a unique audio identity. 5. 𝐄𝐧𝐭𝐞𝐫𝐩𝐫𝐢𝐬𝐞-𝐑𝐞𝐚𝐝𝐲 𝐂𝐨𝐦𝐩𝐥𝐢𝐚𝐧𝐜𝐞 𝐚𝐧𝐝 𝐈𝐧𝐭𝐞𝐠𝐫𝐚𝐭𝐢𝐨𝐧 - Safe for copyright, easy API integration, and ready to fit into your workflows. -- Want to build excellent sounds for your brand? Email us at contact@techificial.ai and see what Stable Audio 2.5 can do for your creative work. #TechificialAI #AudioInnovation #EnterpriseAI #SoundBranding #StableAudio #CreativeAI

こっちで StableAudioに指示をだすとこうなる stableaudio.com/1/share/8e11… もうちょっとスピード感が欲しいところ

👽✨ 東京でひっそり潜むエイリアン！ 🌧️ フォースフィールドの傘 ☂️ で雨を防ぎ、 💫 ホログラムの変装 が光り、ネオン街に溶け込む 🏙️⚡ 🎨 ワークフロー: 🖼️ Qwenモデル Raena (LoRA) 🎞️ Wan 2.2 I2V（最初/最後のフレーム） 🔍 Tensorrt 4x (Animesharp) 🎶 StableAudio：雨、雷、足音の水しぶき、ホロテックの音 🌧️🎛️ @ComfyUI #comfyui #QwenImage #Wan22 #stableaudio #AIart #サイバーパンク

#comfyui #wan22 Upscaled #stableaudio #stablediffusion #hidream Image2Video

Made using @BlocksparkAI , @ComfyUI , and @stableaudio Future Fusion is a completely AI Audio Reactive Music Video Experience . this is Track 5 , Enjoy! #aimusicvideo

Tomorrow morning @dadabots are showing how to compile @stableaudio into a lightweight app for CPU, MacBook, or phone Push your AI music experimentation 👇