- Published on
Google Flow Gets Audio β Veo 3.1 Brings Sound to AI Video Creation
AI video now speaks β and sounds like it.
Google''s AI creative studio Flow has applied Veo 3.1 updates, and the most significant change is audio. Until now, Flow''s core features only generated visuals. Sound has arrived.
What Changed
Here is a summary of the key changes in the Veo 3.1 update.
1. Audio Support β Across All Core Features
Before this update, Flow''s "Ingredients to Video," "Frames to Video," and "Extend" features produced visuals only. All three now generate audio as well.
- Ingredients to Video: Videos assembled from multiple image sources now include background audio.
- Frames to Video: Filling in the motion between a start and end frame now produces matching audio.
- Extend: Extending an existing video now carries the audio forward naturally.
2. High-Resolution Upscaling
Generated videos can now be upscaled to higher resolutions.
| Subscription Plan | Max Resolution |
|---|---|
| Plus, Pro | 1080p |
| Ultra | 4K |
3. Native Vertical Video
9:16 portrait orientation is now supported natively. When using reference images, the vertical format is preserved without automatic cropping. This makes Flow directly usable for Shorts, Reels, and TikTok content.
4. Object Editing Tools
New tools let users select a region of a video and either add or remove an object using a text prompt. This is similar to the "erase and replace" tools familiar from photo editing β now applied to video.
What Is Flow?
Flow is Google''s AI video and image creation studio, launched in early 2026. It unified Google''s generative AI tools β Veo for video, Imagen for images, and Whisk β into a single interface. Projects and assets previously in Whisk and ImageFX moved into Flow as part of the consolidation.
The idea is to handle the entire creative workflow in one place: generate images from text, combine them into video, and edit the result β all without leaving the tool.
Is It Free?
Google offers 50 Flow credits per day to users without a Google AI subscription. Basic features are accessible for free, but high-resolution upscaling and audio generation on longer videos typically require a paid plan.
Why Does Audio Matter?
AI video tools made dramatic progress in 2024 and 2025, but audio always remained a separate step. You would generate the video with AI, then source music or sound effects from a different tool and layer them in post-production.
Flow''s audio integration collapses that two-step process. For short social media content or educational explainer videos, the workflow becomes meaningfully simpler.
Edtech Perspective β Lower Barrier for Educational Content
The biggest friction point in using AI video tools in education had been audio. A short explainer video requires narration recording, background music selection, and editing β all separate from visual generation.
With audio now part of Flow''s generation pipeline, creating a "15-second introduction to today''s lesson" from a few prompt lines becomes a realistic option. The quality is not perfect yet, but the direction is clear.
Tips for Use
- Prepare source images with clean backgrounds and high resolution for best results in Ingredients to Video.
- Use explicit audio prompts like "ambient nature sounds, no music" to steer the generated audio toward your intent.
- For vertical video, specify "9:16 portrait" in your prompt to get the composition you want without cropping.
- With 50 free daily credits, generate low-resolution previews first and save upscaling for final versions.
Sources:
- Bringing new Veo 3.1 updates into Flow to edit AI video β Google Blog
- Google rolls out updates to image and video tool Flow AI β Social Media Today
- Google Veo 3.1: The Next Generation of AI Video β VEED
- Google Flow AI 2026 Announcement: Unified Video Workspace β bonega.ai
- What Is Google Flow? Complete Guide to Google''s AI Filmmaking Tool β simpliaxis.com