- thereach.ai
- Posts
- Adobe teases a new sound AI, Alibaba will compete with Sora, Ideogram competes with Midjourney
Adobe teases a new sound AI, Alibaba will compete with Sora, Ideogram competes with Midjourney
Sorin Ciornei
February 29, 2024
Welcome, AI enthusiasts
Adobe teases a new sound AI, Alibaba will compete with Sora, Ideogram competes with Midjourney
In today’s newsletter:
Adobe previews for its new soundAI solution
Ideogram get $80M in funding so we put it to the test against Midjourney
Alibaba shows EMO, a image to video AI that looks just as good as Sora
Google CEO is upset about the Gemini wokeness blunder
Meta to release GPT4 competitor in July
NEWS OF THE DAY
What: An early-stage generative AI music generation and editing tool, Project Music GenAI Control allows creators to generate music from text prompts, and then have fine-grained control to edit that audio for their precise needs.
Key points:
🎵 Adobe Research is developing a generative AI tool called Project Music GenAI Control for crafting and editing custom audio.
🎶 This tool allows creators to generate music from text prompts and provides fine-grained control for editing audio according to specific needs.
🎚️ Users can input text prompts like "powerful rock" or "happy dance" to generate music, and then perform detailed editing using a simple user interface.
🖼️ Similar to Adobe's Firefly, this generative AI model enables pixel-level control for shaping, tweaking, and editing audio.
Noteworthy: Adobe is adding more capabilities to its AI solutions. This is also in the industries that are likely to be impacted the hardest initially, graphic design and content creation.
EDITORIAL
What: We read a press release where we found out Ideogram received $80M in funding for its image AI. We decided to test it out against Midjourney and see how it matches.
Key points:
🚀 Ideogram 1.0 is the latest text-to-image model introducing transformative capabilities.
🎨 Developed from the ground up, Ideogram 1.0 excels in text rendering, photorealism, and introduces the Magic Prompt.
📈 Ideogram 1.0 claims to set a new industry standard by reducing error rates by nearly 2x compared to existing models.
🌟 Image AI tests reveal user preference for Ideogram's response to prompts over Dalle3 and Midjourney.
Noteworthy: Two conclusions - we tested Ideogram v1 with Midjourney v6. If Ideogram continues to improve, their v6 will be insane. Also, Ideogram does something that Midjourney could never do - create text acurrately.
LATEST NEWS
Audrey Hepburn with the Joker face paint, photography
Alibaba Group has pioneered groundbreaking technology with EMO: Emote Portrait Alive. This innovative project, developed by the Institute for Intelligent Computing at Alibaba, introduces an expressive audio-driven portrait-video generation framework. Let’s look at the capabilities, methods, and diverse applications of EMO.
This certainly gives an awesome new path for people to start bringing their old pictures back to life, like they did in Harry Potter. Check out the article to see the sound since that is half of the achievement.
Google is working to fix its Gemini AI tool, CEO Sundar Pichai told employees in a note on Tuesday, saying some of the text and image responses generated by the model were "biased" and "completely unacceptable".
This shows how far ahead OpenAI are with their LLM, since we already heard rumors OpenAI is planning on releasing GPT5 this year.
MIDJOURNEY PICK OF THE DAY
Paddington and the grand tour, book cover painting by 8 year old kid, red bus
Check out our guides on how to generate the best Midjourney pictures, from fashion, nature, street art or illustration.