Creativity AI #7: Your weekly dose of creativity and productivity
Midjourney Office Hours (2025-02-12), Midjourney documentation website, ElevenLabs Studio, Perplexity $1 million dollar question, Ideogram batch image generation with Chatgpt, Google NotebookLM Plus
Creativity AI is a new publication that brings together interesting articles and recent developments in the AI world related to creativity and productivity.
Midjourney Office Hours (2025-02-12)
Midjourney V7 is still in development, and its release date is unknown. Its new "Omni-reference" feature allows for consistent integration of characters, objects, and logos into the generated images.
Midjourney is still looking into the best way to launch its video feature. It aims to provide a high-quality video product that aligns with the Midjourney aesthetic while making video generation accessible to the masses at an affordable price.
A new folder and workspace system is being developed to simulate Discord's channels, which are often used to generate and organize images for specific projects.
Big batch image generation interfaces may be delayed until V7.
The timeline for real-time 3D features is uncertain due to limited R&D resources.
Midjourney's ultimate goal is to create images, videos, and 3D models in real time.
The current focus remains on improving 2D image generation (better performance, aesthetics, and fewer issues), as well as future "secret" projects.
Several hardware-related projects are currently underway, including the acquisition of a large warehouse to manufacture specialized devices.
Midjourney Refreshed Documentation Website
Midjourney has revamped its documentation website. Take a look at the new website here.
The new documentation site is more than just aesthetically pleasing. It also organizes the information. I like the Full Parameter List page, which gives an overview of the available parameters.
There is also a chat function called "Ask AI" where you can type your questions and get help—a significant improvement in getting quick answers about Midjourney.
It's also nice to see that Midjourney didn't just abandon the Discord users but also provided a comparison table of the features offered by the web versus Discord.
In conclusion, the new documentation site is a significant improvement and beneficial to Midjourney users of all levels. It is an indispensable and authoritative reference for using Midjourney and exploring new techniques that have yet to be discovered.
ElevenLabs
ElevenLabs has made Studio, its text-to-audio editor, available to all users, including free users.
This free tool allows you to create the following:
Audiobooks, articles, podcasts, and voiceovers
Set and adjust pauses for AI voices
Easily assign voices to paragraphs (multiple characters)
You can watch it here to learn more.
Why does this matter?
Studio allows free users to create up to three projects. (Paid users can create an unlimited number of projects).
You can test out ElevenLabs features before subscribing.
Perplexity
Perplexity is giving away $1 million just for asking a question starting on Feb 10th, 2025 at http://pplx.ai/sweepstakes - but only to US residents. Talk about using AI to become a millionaire.
Why does this matter?
Nothing. Unless you’re the lucky winner of that "lottery."
Join me and other crybabies in staring at Perplexity for denying us another chance to win $1 million.
Look, Kitty! Look! That's someone else's $1 million...
Ideogram
Ideogram released a video tutorial for batch image generation using Ideogram Batch and ChatGPT.
Briefly, batch image generation is only available with Ideogram Pro ($48 per month). Users can manually fill out spreadsheet details, such as prompts and parameters, or upload the spreadsheet to ChatGPT for automatic prompt generation. The completed spreadsheet is then uploaded to Ideogram for batch generation.
You can watch the video here.
Why does this matter?
If you need to create a large number of images for marketing or sale on stock photo sites, batch image generation is a good option.
While Ideogram 2.0 excels at creating typography and text-heavy images, it still falls short of Midjourney in terms of image quality, aesthetics, and flexibility (style and image references, mood board, etc.). Thus, if you intend to use the images in situations requiring the highest quality and style consistency, be aware of Ideogram limitations and use Midjourney instead.
I find it interesting to upload a spreadsheet (CSV) template to ChatGPT and request that the bot fill in the details for you, using the spreadsheet structure as a guide and following your additional instructions. It is quite simple yet so effective, but why didn't I think of that?
Inspiration
Recently, scientists discovered a new zombie-spider fungus species, Gibellula attenboroughii, in cave systems across Ireland, infecting two species of orb-weaving spiders.
The fungus appears to manipulate host behavior, causing infected spiders to move from their hiding places to exposed positions on cave ceilings and walls before dying, thereby facilitating spore dispersal. The paper was published on January 25, 2025.
This is what an infected spider looks like in real life.
I'm a big fan of "The Last of Us". Inspired by the above zombie spider, I used Midjourney to create more new zombie spiders, curious to see what the bot could come up with.
The results are quite amusing. Not scary, but somewhat humorous. My Zombie Spiders resemble food-spider hybrids... Ladies and gentlemen, meet the Sushi, Fried Eggs, and Cauliflower spiders!
Pro tip: Style reference (Sref) using a real-life photograph creates realistic photographic fictional creatures.
Google NotebookLM Plus is now available to Google One AI Premium users
Google's AI product names are extremely confusing.
First, there was Google Bard. Then Bard "died" and rebranded as Gemini. Next, we have Gemini Advanced, Gemini Advanced 2.0 Flash, Gemini Advanced 2.0 Pro Experimental, and so on. When I decided to try Google AI and subscribe, I came across a product called "Google One AI Premium" - it's like another product. I had to double-check to make sure I signed up for the right thing.
That’s not all. When I finally get into Gemini Advanced, there is no button to switch to NotebookLM Plus. It's actually on another website. Huh!
If you are new to NotebookLM, you can learn more here.
https://notebooklm.google
You can access the app here.
https://notebooklm.google.com
NotebookLM Plus lets users set up a chat to have a specific conversational style. You can also adjust the bot's response length.
I ask the bot to pretend to be an evil witch and explain spiders to me. I found it more entertaining than the standard chat.
So it works. It's as if I'm speaking with a witch, sort of. The "witch" gives a scientific explanation.
You can also edit the audio to create a podcast. However, there is currently no option for changing the speakers' voices to witches.
The chat settings for bot personality are only for texting and not for changing the audio. It would be ideal if it allowed users to generate voices from ElevenLabs. But not for now.
Why does this matter?
NotebookLM is extremely popular because it makes peer-reviewed scientific publications simple to understand and entertaining through the generated podcast. You can speak with the hosts, and they will answer your questions. You can also create learning materials such as study guides and FAQs. Quite interactive and entertaining.
NotebookLM Plus includes additional features such as chat customization, audio focus areas, analytics, and more, which may be useful if you intend to use it extensively for study or exploration of new topics of interest.
Recent articles
Free Geeky Animals' Sref code: 2026311969
Cover prompt: romance --ar 16:9 --sref 2026311969 --v 6.1
Invitation to publish on Creativity AI
Please contact me if you want to publish on Creativity AI.
This publication will be distributed through Substack and Medium platforms, reaching thousands of readers.
Email: animalsgeeky@gmail.com