MULTIMODAL AI

Consistent Voice Generation Challenge in Multi-Scene AI Creation

By

–

20 May 2025 21h14

So the immediate problem that I see with this is: How to generate multiple scenes with a consistent voice.

You can use elements for consistent visuals, but those voices are going to change, right? I guess a voice element will come, or reference audio from a prior gen? https://t.co/VhC9RYfybb pic.twitter.com/J8oa24sPYX
— fofr (@fofrAI) 20 mai 2025

So the immediate problem that I see with this is: How to generate multiple scenes with a consistent voice. You can use elements for consistent visuals, but those voices are going to change, right? I guess a voice element will come, or reference audio from a prior gen?

→ View original post on X — @lmthang

Google Launches AI Ultra Plan with VIP Access to Advanced Tools

By

@lmthang

–

20 May 2025 21h10

We didn't propose the Google AI Ultra plan $249.99/month, but hopefully it will be well worth it with your VIP access to Google AI (DeepThink, Veo3, Project Mariner, etc) 🙂

Veo 3 Generates Video with Native Spoken Dialogue

By

–

20 May 2025 21h09

Here's that Veo 3 dialogue video, where Veo natively added the spoken words into the scene.

Prompt:
> A medium shot frames an old sailor, his knitted blue sailor hat casting a shadow over his eyes, a thick grey beard obscuring his chin. He holds his pipe in one hand, gesturing… https://t.co/9c5OTXc1RY pic.twitter.com/BhHWUROMQt
— fofr (@fofrAI) 20 mai 2025

Here's that Veo 3 dialogue video, where Veo natively added the spoken words into the scene. Prompt:
> A medium shot frames an old sailor, his knitted blue sailor hat casting a shadow over his eyes, a thick grey beard obscuring his chin. He holds his pipe in one hand, gesturing

→ View original post on X — @testingcatalog

LearnLM integrated into Gemini 2.5

By

@testingcatalog

–

20 May 2025 21h00

LearnLM is now directly integrated into Gemini 2.5 and is being used on Gemini and NotebookLM

DALLE-3 Impressiveness: AI Image Generation Vibes Analysis

By

–

20 May 2025 20h58

It's very impressive but with a dalle3 vibe.

→ View original post on X — @paulroetzer

Video Voice Models and Advanced Reasoning Driving Massive Compute Demands

By

@paulroetzer

–

20 May 2025 20h44

Plus, the ability to serve and scale video + voice, along with advanced reasoning models! That is going to be burning up the compute.

Veo 3 Demonstrates Impressive Capabilities for Creative Prompting

By

–

20 May 2025 20h38

Veo 3 looks incredibly good.
Also, these are my sorts of prompt. pic.twitter.com/Nq5Wxwnbon
— fofr (@fofrAI) 20 mai 2025

Veo 3 looks incredibly good.
Also, these are my sorts of prompt.

→ View original post on X — @testingcatalog

Google unveils Veo 3 with AI video and audio generation

By

@testingcatalog

–

20 May 2025 20h27

BREAKING 🚨: Google announced Veo 3, a new version of its AI video generation model that supports simultaneous video and audio generation!

Veo 3 will be available now in Gemini for Google AI Ultra subscribers in the U.S.

Veo 3 sample 👀 pic.twitter.com/Bvg0CUD1YD
— 🚨 AI News | TestingCatalog (@testingcatalog) 20 mai 2025

BREAKING : Google announced Veo 3, a new version of its AI video generation model that supports simultaneous video and audio generation! Veo 3 will be available now in Gemini for Google AI Ultra subscribers in the U.S. Veo 3 sample