Some observations about Veo 3 generations: – sometimes videos come back without any audio
– it likes to add its own subtitles when there is dialogue
– if you want voice overs or dialogue, you have to prompt for it
– generating videos without audio feels wrong now
– sometimes
MULTIMODAL AI
-
Veo 3 Video Generation: Audio and Subtitle Issues Observed
By
–
-
Veo 3 Generates Award-Winning Anime Scene with Nuanced Dialogue
By
–
A Veo 3 anime prompt: > an anime scene, two mages have a conversation about the dangers of casting spells that undo death, the scene is set in a busy Tokyo cafe, exceptional acting with intensity – award winning animation and performances (definitely not dry and flat)
-
Using Gen-4 References to Extract Objects in AI Workflows
By
–
Use Gen-4 References to extract objects.
-
Native Multi-Speaker Voice Generation
By
–
Native speech generation can produce audio with multiple speakers from text. Includes:
– Gemini 2.5 Flash Preview TTS
– Gemini 2.5 Pro Preview TTS -
Gemini 2.5 Flash: New Audio Models
By
–
BREAKING : Native Speech Generation and Live Audio Generation with Gemini 2.5 Flash is now available on AI Studio! 4 new models from the Gemini 2.5 family Confirmed
-
Stitch: Google Labs Tool Generates Designs from Text to Figma
By
–
8. Stitch: A Google Labs product to generate designs and UIs from text prompts and download them into Figma
— Rowan Cheung (@rowancheung) 21 mai 2025
Coming from Google's acquisition of Galileo AIpic.twitter.com/d8IMXMF3hY8. Stitch: A Google Labs product to generate designs and UIs from text prompts and download them into Figma Coming from Google's acquisition of Galileo AI
-
Google I/O 2025: 11 Most Impressive Announcements Revealed
By
–
It's only been 1 day of Google I/O 2025, and the announcements have already been incredible.
— Rowan Cheung (@rowancheung) 21 mai 2025
The 11 most impressive reveals of I/O 2025 so far:
1. Revamped project Astra with native audio dialogue, UI control, content retrieval, calling, and shoppingpic.twitter.com/y1Gxp5jfD9It's only been 1 day of Google I/O 2025, and the announcements have already been incredible. The 11 most impressive reveals of I/O 2025 so far: 1. Revamped project Astra with native audio dialogue, UI control, content retrieval, calling, and shopping
-
15 Game-Changing AI Tools For Social Media Content Creation
By
–
15 Game-Changing AI Tools For Social Media And Content Creation Discover the most powerful #generativeAI tools that are transforming how businesses approach #socialmedia management and #contentcreation. From #video generators like #CapCut to comprehensive #socialmediaplatforms
-
Gemma 3n: Multimodal AI Model Running on 2GB RAM
By
–
Meet Gemma 3n, a model that runs on as little as 2GB of RAM 🤯 It shares the same architecture as Gemini Nano, and is engineered for incredible performance. We added audio understanding, so now it’s multimodal, fast and lean, and runs on-device (no cloud connection required!) pic.twitter.com/2FyzJHVGZa
— Google AI (@GoogleAI) 20 mai 2025Meet Gemma 3n, a model that runs on as little as 2GB of RAM It shares the same architecture as Gemini Nano, and is engineered for incredible performance. We added audio understanding, so now it’s multimodal, fast and lean, and runs on-device (no cloud connection required!)
-
Google Introduces Gemini Diffusion: Faster Non-Sequential Text Generation
By
–
Today we introduced Gemini Diffusion⚡️ (& DeepThink, Veo3, Imagen4, 2.5 updates…).
— Oriol Vinyals (@OriolVinyalsML) 20 mai 2025
It's been a dream of mine to remove the need for "left to right" text generation. It's so fast, that we had to *slow down* the video during the presentation.https://t.co/YW73ZPqJ5a pic.twitter.com/N3BXZs9ELeToday we introduced Gemini Diffusion (& DeepThink, Veo3, Imagen4, 2.5 updates…). It's been a dream of mine to remove the need for "left to right" text generation. It's so fast, that we had to *slow down* the video during the presentation. https://
deepmind.google/models/gemini-
diffusion/
…