Meta’s AudioCraft is AudioGen + MusicGen + EnCodec
Plus: Google Lab Sessions, LLaMA2-Accessory: Toolkit for LLM Development.
Hello Engineering Leaders and AI Enthusiasts!
Welcome to the 77th edition of The AI Edge newsletter. This edition brings you Meta’s AudioCraft, a single-stop generative AI model for your audio needs.
And a huge shoutout to our incredible readers. We appreciate you! 😊
In today’s edition:
🎵 Meta’s AudioCraft is AudioGen + MusicGen + EnCodec
💡 Google’s latest venture to continue experimenting with AI
🚀 LLaMA2-Accessory: An Open-source Toolkit for LLM Development
🧠 Knowledge Nugget: Generative AI: Why Artists Need Not Worry by
Let’s go!
Meta’s AudioCraft is AudioGen + MusicGen + EnCodec
Meta has introduced AudioCraft, a new family of generative AI models built for generating high-quality, realistic audio & music from text. AudioCraft is a single code base that works for music, sound, compression & generation — all in the same place. It consists of three models– MusicGen, AudioGen, and EnCodec.
Meta is also open-sourcing these models, giving researchers and practitioners access so they can train their own models with their own datasets for the first time. AudioCraft is also easy to build on and reuse. Thus, people who want to build better sound generators, compression algorithms, or music generators can do it all in the same code base and build on top of what others have done.
Why does it matter?
AudioCraft is a significant step forward in generative AI research. It opens up unprecedented possibilities for creating unique audio/music– whether for video games, merchandise promos, YouTube content, educational purposes, etc. Moreover, the open-source initiative will further help advance the field of AI-generated audio and music.
(Source)
Google’s latest venture to continue experimenting with AI
Google has announced Lab Sessions, a series of experimental AI collaborations with visionaries – from artists to academics, scientists to students, creators to entrepreneurs. Here’s a Google Labs Session exploring how AI computer vision models could help people learn sign language in new ways.
This will help Google continue to experiment with AI. And it aims to showcase its existing and future collaborations across all kinds of disciplines.
Why does it matter?
In today's AI landscape, there is a plethora of ongoing developments, hardware advancements, and new research. But it is crucial to ensure that AI technology is put in the hands of actual people to solve real-world problems. Google’s initiative can ensure AI is practical, user-centric, and positively impacts society.
(Source)
LLaMA2-Accessory: An Open-source Toolkit for LLM Development
LLaMA2-Accessory is an advanced open-source toolkit for pre-training, fine-tuning, and deployment of Large Language Models (LLMs) and multimodal LLMs. Its repository is mainly inherited from LLaMA-Adapter with more advanced features.
Thus, it supports more datasets, tasks, visual encoders, and efficient optimization methods. (LLaMA-Adapter is a lightweight adaption method to efficiently fine-tune LLaMA into an instruction-following model).
Why does this matter?
It will allow to easily and quickly experiment with and build upon state-of-the-art language models, saving time and resources in the development process. Moreover, its open-source nature democratizes access to advanced AI tools, enhancing engagement and progress toward groundbreaking AI solutions across various industries and domains.
Knowledge Nugget: Generative AI: Why Artists Need Not Worry
Generative AI has been making remarkable progress– generating writing akin to Shakespeare, music emulating any artist's voice (dead or alive), and artwork in the style of renowned artists like Michelangelo. Thus, it may be a nightmare for artists due to:
Content creator's/artist’s trademark violation or copyright infringement
Content consumers are misled/duped by AI content
In this article,
proposes a possible solution of using verified accounts (Blue CheckMarks) to distinguish original content from AI-generated content. It also talks about future challenges, but despite the concerns, the article emphasizes that the future will be powered by AI, presenting both challenges and exciting opportunities.Why does this matter?
The article addresses the crucial concerns surrounding the impact of Generative AI on artists and content creators by offering a practical approach. Further, it encourages discussions on how to navigate the evolving landscape of AI responsibly and creatively.
What Else Is Happening❗
🔍Instagram is working on labels for AI-generated content (Link)
🌐Google’s generative search feature now shows related videos and images (Link)
📸Tinder tests AI photo selection feature to help users build profiles (Link)
🤖Alibaba rolls out open-sourced AI model to take on Meta's Llama 2 (Link)
🚀IBM and NASA announced the availability of the watsonx.ai geospatial foundation model on 🤗 (Link)
📈 Thursday Trajectory
In the spotlight today: Inworld AI
They recently raised $50M at a $500M valuation and are backed by various tech giants like Meta, Microsoft, Intel, LG, and more.
Inworld AI develops artificial intelligence technology to create more realistic and interactive non-player characters (NPCs) for video games.
Their "Character Engine" uses advanced AI models to make game characters more lifelike by giving them cognition and memory.
Their offering can be integrated with gaming engines like Unity and Unreal and this might be the biggest advantage for Inworld AI.
They won’t compete but facilitate all the major players in the gaming industry. Thus the growth potential is huge for this one.
🛠️ Trending Tools
Phrasion: AI blog writer and article generator for more effective content production.
NeuralBox: AI Memory Extension to remember anything you see using photos.
Coleap: Deliver great experiences to your users with little operational effort powered by AI.
Skill AI: Create personalized learning paths for any skill with progress tracking.
Dash AI: Instant access to ChatGPT on every webpage for increased productivity.
Voicemy AI: Clone voices, train AI models, compose melodies and share your passion.
Taja AI: AI-Powered SEO expert for YouTube content creators to assist with publishing.
Blizzy: Secure AI assistant for chat, document search, and online browsing with privacy and security.
That's all for now!
Be in the company of industry frontrunners! Subscribe to The AI Edge and join the ranks of respected readers from Moody’s, Vonage, Voya, WEHI, Cox, INSEAD, and other notable organizations.
Thanks for reading, and see you tomorrow. 😊
I love the idea of using verified accounts to distinguish original content from AI-generated ones. It's a practical approach to address concerns and foster responsible AI use.