sayzardGoogle AI Studio (@GoogleAIStudio)
Google AI Studio에서 Lyria 3 Pro Preview를 직접 체험할 수 있는 새 음악 생성 페이지를 안내한다. 최신 음악 생성 모델을 테스트해볼 수 있는 공개 미리보기로 보인다.
nttluke.nft (@NTTLuke)
autoresearch-mlx를 변형해 작은 GPT를 ABC 표기법의 포크 멜로디로 학습시키는 프로젝트를 소개했다. 로컬 환경에서 실행되는 연구 루프, 경량 UI, 실행 비교, 라이브 모델 테스트를 결합해 소규모 음악 생성 연구를 쉽게 실험할 수 있게 했다.
nttluke.nft (@NTTLuke) on X
https://t.co/jzaTtkDrVT Built a small twist on autoresearch-mlx: Adapted the loop to train tiny GPT on folk melodies in ABC notation with lightweight UI to track runs, compare winners, and test the best model live. Local researcher brains run through @AgnoAgi + @lmstudio .
KloudiHub🎵 Google just dropped Lyria 3 Pro — and it can write full songs inside Gemini.
Multi-minute. Structured. Production-ready.
Generative music isn't a toy anymore.
Wes Roth (@WesRoth)
Suno 5.5가 공개한 새로운 음악 생성 능력이 화제를 모은다. 사용자는 직접 만든 곡의 라이브 반응을 공유하며, 몽골 목노래와 덥스텝을 결합한 결과가 인상적이라고 평가했다.
AZIZ | AI (@aziz4ai)
음악 생성 AI 'Suno'가 음성 복제 기능을 통해 10초 음성 샘플로 화자 확인 후, 사용자의 목소리로 원하는 가사를 자연스럽게 노래하게 만드는 기능을 선보였다고 소개한다. 음악 제작과 보컬 합성의 활용 가능성을 강조한다.
AZIZ | AI 🇸🇦 (@aziz4ai) on X
اليوم اقول نموذج الذكاء الاصطناعي سونو سواها وكسر مجاديف اي نموذج موسيقي مع خاصية الاستنساخ الصوتي ❤️🔥 الفكرة سجل ١٠ ثواني بصوتك، بعدها تاكيد الهوية الصوتية، بعدها تقدر تسوي اي كلمات بصوتك بسهولة وكل اغنية تطلع بصوتك بدقة 👌✨ تجربتي اغنية ستظل عيني (كلماتي، الحاني، وبصوتي) 😍
Google Gemini (@GeminiApp)
Lyria 3 Pro가 오늘부터 Google AI Plus, Pro, Ultra 사용자에게 순차 배포됩니다. Gemini 웹과 모바일 앱에서 'Create Music' 도구와 모델 선택 기능을 통해 바로 사용할 수 있습니다.
Google for Developers (@googledevs)
Lyria 3와 Lyria 3 Pro가 Gemini API와 Google AI Studio에서 퍼블릭 프리뷰로 제공되기 시작했다. 두 가지 변형으로 생산성과 지연 시간 요구에 맞춰 선택할 수 있으며, 모델을 더 세밀하게 제어할 수 있는 음악 생성 AI 관련 업데이트다.
Google for Developers (@googledevs) on X
🎵 Lyria 3 and Lyria 3 Pro are now available in public preview via the Gemini API and in @GoogleAIStudio — and it’s music to our ears 🎵 🎼 Choose from two distinct variants to match production & latency needs (Lyria 3 Pro and Lyria 3 Clip) 📢 Direct the model with more
Harold Sinnott 📲Just a month after dropping Lyria 3, @Google is already turning up the volume! 🎛️
They just announced Lyria 3 Pro from @GoogleDeepMind , their most advanced AI music model yet. Creators can now build tracks up to 3 minutes long with way more creative control. Plus, they’re integrating it into more Google products starting today. 🎶✨
Ready to make some music?
@GeminiApp @Google.gemini #AI #earlyaccess #Lyria3Pro #MusicAI
PsychoticSheepJust ran Demucs completely locally on my system (RX 6700 XT / 16 GB RAM).
Demucs is an open source AI model for music source separation, developed by Meta. It can split a full song into individual stems like vocals, drums, bass, and other instruments, making it useful for remixing, transcription, and audio analysis.
Test track: Fear of the Dark by Iron Maiden
(https://www.youtube.com/watch?v=bePCRKGUwAY)
Setup:
- Demucs installed via pip
- Model: htdemucs (default)
- Input converted to WAV using ffmpeg
- GPU acceleration via ROCm
Setting it up is tricky because Demucs is tightly pinned to older PyTorch versions, so you have to install dependencies manually and use "--no-deps" to avoid breaking your (ROCm-)PyTorch setup.
Result:
Very clean vocal separation in most parts. Some artifacts appear during very loud or distorted sections (e.g. emotional peaks or shouting).
Next steps / possibilities:
- Normalize and filter audio before separation
- Extract vocals for transcription or remixing
- Create karaoke / instrumental versions
- Combine with Whisper for lyrics
- Batch processing for datasets
- Model: htdemucs_ft (higher quality)
Video workflow:
- Recorded with OBS
- Edited in Kdenlive
- Transcoded with VAAPI (H.264)
No cloud, real hardware.
Everything runs on Linux, so anyone can set this up.
Works on CPU as well, but much slower.
#Demucs #AI #MachineLearning #AudioSeparation #MusicAI #OpenSource #Linux #ROCm #AMD #DeepLearning #AudioProcessing #Vocals #Karaoke #StemSeparation #SelfHosted #NoCloud #FOSS #Tech #LocalAI #MetaAI
Demucs is an open source AI model for music source separation, developed by Meta. It can split a full song into individual stems like vocals, drums, bass, and other instruments, making it useful for remixing, transcription, and audio analysis.
Test track: Fear of the Dark by Iron Maiden
(https://www.youtube.com/watch?v=bePCRKGUwAY)
Setup:
- Demucs installed via pip
- Model: htdemucs (default)
- Input converted to WAV using ffmpeg
- GPU acceleration via ROCm
Setting it up is tricky because Demucs is tightly pinned to older PyTorch versions, so you have to install dependencies manually and use "--no-deps" to avoid breaking your (ROCm-)PyTorch setup.
Result:
Very clean vocal separation in most parts. Some artifacts appear during very loud or distorted sections (e.g. emotional peaks or shouting).
Next steps / possibilities:
- Normalize and filter audio before separation
- Extract vocals for transcription or remixing
- Create karaoke / instrumental versions
- Combine with Whisper for lyrics
- Batch processing for datasets
- Model: htdemucs_ft (higher quality)
Video workflow:
- Recorded with OBS
- Edited in Kdenlive
- Transcoded with VAAPI (H.264)
No cloud, real hardware.
Everything runs on Linux, so anyone can set this up.
Works on CPU as well, but much slower.
#Demucs #AI #MachineLearning #AudioSeparation #MusicAI #OpenSource #Linux #ROCm #AMD #DeepLearning #AudioProcessing #Vocals #Karaoke #StemSeparation #SelfHosted #NoCloud #FOSS #Tech #LocalAI #MetaAI
Kyoko KitamuraThe "AI" musical group Neon Oni is set to perform live concerts in Japan. "Initially, Neon Oni existed solely in the digital realm. Its creator, known as Kage, used Suno AI to generate music and conceived seven fictional members from Tokyo." #music #AI #AImusic #MusicAI
forklog.com/en/ai-band-n...
AI Band Neon Oni to Tour Japan...
AI Band Neon Oni to Tour Japan...
