AI · AIC-112

AI Video Creation with Google Flow

AI Video Creation with Google Flow is a hands-on course in making video with Google's own tools, from characters created with Nano Banana and clips generated from text, frames and reference images in Flow to dialogue, longer scenes, credit planning, editing and publishing. It suits content creators, marketing teams and online sellers, and you leave with a short ad and a social cut-down.

Updated 13 Nov 2026

From 6,210 THB / person 6,900 −10% excl. VAT 7% · group rates available

Corporate training quote Join the waiting list

PDFDownload the course outline

Duration12 hours · 2 days
FormatOnsite / live online
Next roundOn request
CertificateIncluded

Course overview

Google now brings its AI video tools together in one place: Google Flow, built for multi-shot projects; the Gemini app, which creates clips from a chat; the Veo and Gemini Omni video models, which generate pictures with speech built in; and the Nano Banana image models for characters and frames. Many people try them and end up with characters whose faces change in every shot, scenes that jump, and credits that run out before anything usable is made.

This course is a deep dive into Google's own tools from start to finish. Learners plan shots and a credit budget, create characters and product images with Nano Banana, and generate clips with Text to Video, Frames to Video and Ingredients to Video. They keep characters consistent with multi-character dialogue, build longer scenes with Extend and Scenebuilder, and use the Flow Agent to plan and produce variations. They then edit and make versions for YouTube and social media, with guidance on copyright, consent and SynthID. The course closes with a workshop that produces a complete ad or UGC-style clip. (2 days, 6 hours per day, 12 hours in total, Beginner to Intermediate level.)

What you’ll gain

Choose between Flow and the Gemini app, and pick a video model that fits the job and the credits available
Plan a shot list and credit budget before generating, with help from the Flow Agent
Create characters, product images and reference frames with Nano Banana
Generate clips with Text to Video, Frames to Video and Ingredients to Video
Keep characters consistent and create dialogue between several characters
Build longer scenes with Extend and arrange sequences in Scenebuilder
Edit, caption and make versions for YouTube and social media
Check copyright, consent, AI content disclosure and PDPA before publishing

Who this course is for

Content creators and YouTubers who want to make short clips and story scenes with AI
Marketing teams and social media admins who regularly produce ads and UGC-style clips
Business owners and online sellers who already use Google AI and want to make their own product clips
People who have tried video in Flow or Gemini and want results they can control

Prerequisites

Basic computer skills, including handling image and video files
No video editing or filming experience required
The learner's own personal Google account, aged 18 or over
A Google AI plan with enough Flow credits is recommended before the training day

Curriculum

Course Details

A 2-day hands-on course, 6 hours per day (12 hours in total, 09:00-16:00), at Beginner to Intermediate level with no video editing experience needed. It uses Google Flow, the Gemini app, the Veo and Gemini Omni video models and the Nano Banana image models through the learner's own Google account; the full feature set and enough credits for real work require the learner's own

paid Google AI plan. The course is a deep dive into Google's tools only, from characters, dialogue and longer scenes to credit planning, editing and publishing, unlike AI Video Generation for Business, which combines several tools. Learners take home a shot list and credit budget template, a character and video prompt set, the ad or UGC clip and social cut-down made in the workshop, and a pre-publication checklist.

Day 1 Planning, Characters and Generating Clips in Flow

Section 1: Getting to Know Flow, the Gemini App and Google's Video Models

How Flow for multi-shot projects differs from the Gemini app for quick clips from a chat
Video models in Flow such as Veo 3.1 Lite and Gemini Omni Flash: length, resolution and built-in audio
Google AI plans and credits: daily credits, monthly credits and buying more
Features available only on the desktop web app, plus age and regional requirements
Lab: create your first Flow project and explore the prompt box, settings and credit balance

Section 2: Lab: Planning Shots and a Credit Budget with the Flow Agent

Use the Flow Agent and Gemini to break a brief into a script, storyboard and shot list
JSON-style structured prompts that separate subject, action, camera, lighting and audio
Draft at low resolution first, then generate and upscale only the shots that pass
Set the Agent to ask before spending credits, and work out credits per shot before generating
Lab: build a five-shot ad shot list with prompts and a credit budget

Section 3: Lab: Characters and Reference Images with Nano Banana

Choose a Nano Banana model in Flow and the Gemini app according to your plan
Multi-angle character sheets that fix face, hair, wardrobe and signature props
Product, set and style frames to reference in every shot
Use fictional characters by default; real people only with their consent
Lab: create two main characters and product images and save them as a collection

Section 4: Lab: Text to Video and Visual Language for Prompts

Shot sizes, camera angles, camera moves, lenses and lighting that the model understands
Writing dialogue, ambient sound and sound effects into the prompt
Changing one variable at a time instead of regenerating everything
Common failure points: hands, text, logos and unnatural movement
Lab: create the opening shot and compare results across models and resolutions

Section 5: Lab: Frames to Video and Ingredients to Video

Frames to Video: control movement with a first frame, or a first and last frame
Ingredients to Video: combine characters, products and style as references in one shot
Laying out still keyframes for the whole story before turning them into video
Checking that faces, wardrobe and product labels stay consistent across shots
Lab: create character and product shots from the images made in Section 3

Day 2 Dialogue, Longer Scenes, Editing and Publishing

Section 6: Lab: Consistent Characters and Multi-character Dialogue

Use the same reference set in every shot so characters keep their faces
Define each character's voice, such as age, tone and speaking pace
Write two-person dialogue that fits the clip length and lip-syncs well
Testing Thai dialogue, with a fallback plan of dubbing it afterwards
Lab: create a UGC-style product review scene with two characters talking

Section 7: Lab: Longer Scenes with Extend, Scenebuilder and Clip Editing

Extend a clip from its ending so the scene runs longer without jump cuts
Edit an existing clip by instruction, such as changing the background, lighting, product or aspect ratio
Scenebuilder: arrange clips, trim them and preview the whole sequence
Generate several versions at once to choose a hook, and upscale only the shots you use
Lab: join the main scenes into a 20-30 second sequence

Section 8: Lab: Editing and Versions for YouTube and Social Media

Download clips from Flow and arrange them to the script in a video editor such as CapCut
Add Thai captions and music, and balance levels so dialogue stays clear
Make 9:16 cut-downs for Shorts, Reels and TikTok, and 16:9 for YouTube
Thumbnails, titles, descriptions and labelling the use of AI
Lab: edit the main clip and a 15-second short version

Section 9: Copyright, Consent and Transparency for AI Content

AI video is not automatically free of copyright; check Google's terms and your plan
Never impersonate real people, celebrities or other brands; real faces and voices need written consent
SynthID, the invisible watermark in Google AI output, and checking for it in the Gemini app
Labelling AI content under YouTube, TikTok and Meta policies
PDPA for customer and staff images, and a pre-publication checklist

Section 10: Workshop: A Complete Ad or UGC Clip

Workshop: take a brief and plan the shot list and credit budget
Create characters and products with Nano Banana, then generate shots with Frames or Ingredients
Add dialogue, extend scenes and edit with captions
Export the main clip and a social cut-down, review against the checklist and present the work

Schedule & training options

For individuals — public rounds

No public rounds are open right now. Join the waiting list and we will contact you first when the next round opens, or ask us on LINE. Or call 02-570-8449 or 088-807-9770

Join the waiting list Ask about rounds

For organisations — in-house / private

Tailor the content to your team’s tools and projects
Your dates, at your office or live online
Quotation with tax ID for procurement

Corporate training quote

Instructors

Samit KoyomFounder & CEO, IT Genius Engineering

Frequently asked questions

How does this course differ from AI Video Generation for Business?

AI Video Generation for Business covers business video production with several tools together, such as Veo, Kling, ElevenLabs and HeyGen. This course is a deep dive into Google's tools only: Veo and Gemini Omni in Flow and the Gemini app, the Nano Banana image models for characters and frames, consistent characters, dialogue, longer scenes and credit planning, followed by editing and publishing. It suits people who mainly use Google AI and want to go further with it.

Do I need a paid Google AI plan before the course?

Learners use their own Google account and their own credits; the institute does not provide accounts or credits. Without a plan, Flow gives a limited daily credit allowance and basic features only, while longer clips, Extend and upscaling need a paid Google AI plan. We recommend subscribing to a plan with enough credits before the training day. Plan and credit costs are the learner's own, separate from the course fee, and Google's terms may change.

The title says Veo, so why does the course also cover Gemini Omni?

Google updates the models in Flow regularly. Many Flow features now run on the Gemini Omni family alongside Veo, and some versions are being retired. The course therefore teaches methods that carry across versions, such as shot planning, character control and credit budgeting, and uses whichever models Flow offers on the training day.

Can I use Google Flow in Thailand, and can it create Thai dialogue?

Thailand is on the list of places where Google Flow is available, for users aged 18 or over, although some features and voice options may be limited by region. In the course we test Thai dialogue in practice and show a fallback of adding captions or dubbing during editing.

Can I use the videos I create commercially?

AI video is not automatically free of copyright. Your rights depend on Google's terms, the plan you use and whatever you supply as reference material. The course covers checking the terms, avoiding other people's content and a pre-publication checklist. The institute does not guarantee that any output is free of copyright, and higher-risk work should be reviewed by your organisation's legal team.

Can I use my own face or voice, or those of staff or celebrities, in the videos?

Use only your own face and voice, or those of people who have given written consent, and handle their data under PDPA. Never create videos that impersonate real people, celebrities or other brands. Google AI output carries an invisible SynthID watermark, and platforms such as YouTube, TikTok and Meta require realistic AI content to be labelled; the course explains how to comply.