Google Veo 3.1: Audio-First Control for Flow Creators
Flow's growth sets the stage
Flow, Google's Veo-powered filmmaking tool, has already inspired more than 275 million generated videos in its first five months of availability.1 That usage highlighted two recurring creator requests: finer-grained artistic control and richer audio support across existing templates.
Audio-first upgrades inside Flow
Veo 3.1 elevates Flow from silent concept generator to sound-aware storyteller. Google now layers generated audio across Flow's core tools while improving prompt adherence and visual realism:
- Ingredients to Video accepts multiple reference images so that characters, props, and style remain consistent while Veo 3.1 scores the scene with contextual audio.1
- Frames to Video stitches a starting and ending frame into one cinematic transition, now with sound design that follows the action.
- Extend grows sequences to a minute or longer by analyzing the final second of each clip and composing a seamless continuation with matching ambience.
All three modes are still experimental, and Google says it is iterating based on creator feedback as the audio pipeline matures.
Editing flexibility built in
Great ideas rarely ship on the first take, so Flow now exposes editing primitives powered by Veo 3.1:
- Insert lets you drop in any new object or character, complete with lighting and shadows that match the rest of the shot.
- Remove is rolling out soon to cleanly delete unwanted elements while reconstructing the background.2
Together with the upgraded audio stack, these controls nudge Flow toward an end-to-end previsualization and production workflow instead of a one-off prompt editor.
Where to use Veo 3.1 today
Creators can already explore Veo 3.1 inside Flow, while developers and enterprises gain access through the Gemini platform stack:
- Gemini API: exposes the latest video capabilities (including Ingredients to Video, First and Last Frame, and Scene Extension) for custom apps.2
- Vertex AI: offers managed infrastructure for Veo 3.1, with Scene Extension rolling out soon for production workloads.2
- Gemini app: surfaces Veo-powered storytelling features to end users without leaving the consumer interface.
By unifying Flow's creative interface with API and enterprise endpoints, Google is positioning Veo 3.1 as its state-of-the-art engine for audio-rich, prompt-aware video generation.