Create Stunning Videos Effortlessly: Produce professional-quality videos with lifelike AI voices, matching footage, and music seamlessly integrated into SimGate Studio.
AI Video Overview
Easily Generate Videos: Transform text inputs into high-quality videos in no time. Creators select images and scenes for their videos and add voice-over narration using voices from the AI voice library.
Save Time, create videos in just minutes
One tool, integrated into SimGate Studio
Simple To Use, powerful A.I. does the work for you
Professional Quality, millions of video clips or upload your own
Easy to maintain, make updates whenever needed
Want to see what's possible? CLICK HERE to see a video made using Pictory.
Example Use Cases
Create dynamic introduction videos or business cases to set the stage for your learners within your simulations and elearnings.
Showcase topical information and share knowledge through engaging videos to help support learner retention.
Cater to different learning styles and accessibility needs with multi-media content, enhancing overall learning experiences.
Personalize videos to roles, regions, or industries by selecting voiceovers and visual elements, ensuring alignment with specific learning objectives.
General Video Creation Workflow
Initialization (default state of a newly created template)
Transcript - You can paste/type in text that uses AI to generate your scenes and click “Generate scenes“.
For example: “How to play golf.“ or “How to identify a new city to move to.“
Alternatively, you can manually create scenes by clicking the “+“ button in the scenes section of the editor to create your scenes yourself.
Generate Preview (you must have 1 or more scenes in your video)
You have 1 or more scenes in your video (max 20 scenes) you will be able to click “Generate preview”.
You must generate a preview before the final video can be generated
NOTE: “Generate video” will be locked until you do so.
Voice-over generation occurs during this phase
If a scene doesn’t have an asset linked to it an asset will be assigned during the preview generation.
Generate Video (you have generated a preview)
Once you have generated a preview, you can now generate the final video by clicking “Generate video”.
DIY STEPS
Using Pictory in Pagebuilder
Click on AI Generated Video.
Click on the video placeholder, then click on the pen icon to edit.
Enter your script, and then click on Generate preview.
You’ll see your script is broken out into different screens. If you scroll towards the bottom of the edit panel, you’ll see where you can select the voiceover voice.
At each scene, click on the video icon to select the image for the scene. You can choose one of your own, or select from a wide range of stock images.
When you’re ready, you can generate your video.
FAQ
Where is AI Video available within Studio?
The following pages/templates will support AI Generated Video creation:
Conversation
Video Response
Page Builder
Slideshow
Add-on Builder
These will not support AI Generated Video creation:
COA
Perspectives (Add-on)
What configuration options do I have for AI Videos?
Preview Section
Video Player
Pause/Play
Progress bar (shows scenes and video timeline with position indicator)
Ability to reorder scenes using drag-and-drop
Actions Section (unlabeled, does not contain a section header)
Generate/re-gen preview - regenerate the video preview
Generate video - regenerate the final video
Scenes Section
Add scene - click “+“ icon
Collapse/expand - collapses or expands the scenes section
Scene
Toggle voiceover - turn voiceover on/off for that scene (mute/unmute icon)
Asset picker - select an asset for the scene (movie icon)
Delete scene - removes the scene from the video (trash icon)
Asset preview
Video - if video is chosen it will show the video player
Image - if image is chosen it will show the image
Scene text - the text that will be used to generate the voice over for that scene
Voice Over Section
Voice actor - the selected voice actor that will be used to generate the voice over
Clicking the microphone button will launch the AI Audio voice picker
Note: It will show loading spinner in place of the microphone when generating or loading data (voices etc)
Required Section
Required toggle - indicates if the video is required
Enabling this will toggle the prevent seek property in video options since they are correlated
Video Options Section
Preload video - indicates if the video should be preloaded once mounted
Autoplay video - indicates if the video should autoplay once loaded
Loop video - indicates if the video should loop
Prevent seek - indicates if prevent seek should be disabled
Enabling this will toggle the required property in the required section since they are correlated
Show video controls - indicates if the video controls should be shown
Allow download - indicates if the download features is available on the video player
Width Section
Width - Allows the author to change the width of the UI/video element
Align Item Section
Indicates the alignment of the item relative to the page builder column based on the column settings
What does Visible, Hidden, & Disabled mean in the Editor Elements?
Always visible
Scenes, voiceover, width and align item sections
No scenes
Actions section that contains the “Generate/Re-gen preview” and “Generate video” buttons hidden.
Transcript section visible
Video has 20 scenes
Add scene button disabled (20 scenes max allowed)
1 or more scenes
Actions section containing “Generate/Re-gen preview” and “Generate video” buttons visible.
Transcript section hidden
No preview generated
Preview, required & video options sections hidden
Generating scenes (AI), generating preview or generating video
All actions in scenes (except collapse) disabled
“Generate/Re-gen preview” and “Generate video” buttons disabled
Preview progress bar/timeline drag-and-drop and actions disabled (if section is visible)
Voiceover section + voice actor disabled
How do I use the Scene Asset Picker to choose my clips/scenes?
Each scene in the scenes section has a movie icon. Clicking it will launch the asset picker for the scene. There are three sections available:
Stock assets - A library of stock assets that can be searched using the external Pictory service.
Search field - Text input field for the keyword of your search
Search Mode - Video or Image
Search categories - All categories, Agriculture and Forestry, Arts and Entertainment, Autos and Vehicles, Beauty and Fitness, Science/Biological Sciences, Books and Literature, Business and Industrial, Computers and Electronics, Finance, Food and Drink, Games, Health, Home and Garden, Internet and Telecom, Jobs and Education, People and Society, Pets and Animals, Real Estate, Science, Sports, Travel
Search button - Search for assets based on search field, mode and category
Videos - The standard asset picker section for videos in your project. You can upload your own videos for use in your scene.
Images - The standard asset picker section for images in your project. You can upload your own images for use in your scene.
To load more search results from the stock assets section, scroll to the bottom of asset picker.
A loading message will be displayed when stock assets are being loaded. If assets failed to load an error message will be shown.
If no results are located based on your search a “ No results for ${term}” message will be displayed.
What do the various alert messages mean? (e.g. idle, error, generating, etc...)
Idle (no active work is occurring)
Error (error occurred indicating the user must try again)
Generating scenes (AI is creating your scenes)
Generating preview (preview is being actively generated)
Generating video (video is being actively generated)
What are the video specs, limitations, & constraints?
Transcript (AI text-to-scenes) has a 4000 token limit.
If you paste/type text that has more than 4000 tokens, it will prevent you from being able to “Generate scenes” and show an info message indicating the error. Note, tokens are different than text length. We use a “p50k_base” implementation since we are using OpenAI’s “text-davinci-003“ model.
Voice over - The text from all scenes is combined into one string to generate the voiceover. Eleven Labs only supports a text length of 5000 characters. If your combined scenes text is more than 5000 characters, an info message will be shown indicating that the text will be trimmed to meet that requirement.
English is currently only supported by Pictory.
Eleven labs (voice over) does support 29 languages, but Pictory doesn’t support anything but english currently. There is a language detection layer/validator built in to the editor that will detect the language of the combined script text. If English is not the detected language it will prevent you from being able to generate the preview/video.
Max allowed scenes is 20. This number is to prevent the preview progress bar from becoming unmanageable/unusable.
It is recommended that you use videos with a 16:9 aspect ratio @ 1920x1080 for optimal quality.
This template generates videos in 16:9 aspect ratio. If you add assets to a scene that are not 16:9 aspect ratio, the added video will be centered vertically/horizontally and positioned to fit in the 16:9 frame.
If you choose to use your own video/image assets that you upload through the asset picker, these are the + options:
Images -
.png.jpg.jpegVideo -
.mp4
Scenes that are empty and don’t have any text are not considered valid and will automatically assign “Empty scene“ to that scene when generating preview/video. The scene will still remain empty in the editor panel but the voice over will say “Empty scene”.
When can I copy an AI Video?
When using the copy/paste feature in pagebuilder, slideshow, and add-on builder note the following as it pertains to AI-Text-to-Video:
AI Videos cannot be copied when:
Generating Scenes
Generating Preview
Generating Video
AI Videos can be copied when:
Scenes haven’t yet started generating
After scenes have generated but before the preview has started generating
After video generation completion
What is the Generation Process for AI Videos?
Anytime any scene text changes, the order of scenes changes, the voice actor changes or the number of scenes changes it requires the video preview to be regenerated.
The preview must be generated before the final video can be generated.
A unique fingerprint for both the preview and the final video is tracked so you can only generate the preview and final video when necessary.
This is done to prevent Studio authors from unnecessarily burning API minutes and generating previews and final videos when not needed.
Video Preview Asset Preloading
The video preview that appears in the preview section of the editor is a unique video player.
While it may appear that the preview video player is a single video, under the hood that isn’t the case.
Let take this example: You have a video with 10 scenes. Those 10 scenes are actually 10 individual videos/images under the hood + the voice over audio for the entirety of the video. Our custom preview player has to seamlessly combine all of those 10 videos + the voice over together and coordinate the scene timings together to make it appear as one complete video to the end user.
This unique requirement means all assets (videos/images/audio) need to be preloaded before the author is allowed to view the preview. Without this, the preview could break/have timing and synchronization issues.
When you open the editor for the first time or if you add, replace or update any of the video assets (including regenerating the preview), it requires all new assets to be loaded.
When this is occurring, you will see a message that says “Preparing preview“. That message will be displayed until all assets required to play the preview have finished loading.
Failure to Load: In the event that 1 or more of the scene assets fails to load, the pre-loader will attempt to reload the asset(s) continuously every 10 seconds.
If this occurs, you will see an error message in the preview player that says “Failed to load media”. Below that it will also show a message that begins a countdown that says “Retrying in 10 seconds“. That message will count down to 1 and then say “Retrying now“.
A similar message will be displayed on the scene(s) indicating which asset(s) has failed to load.
The editor will automatically scroll to the first asset that failed to load to indicate to the user which asset is failing to load. If the pre-loader successfully loads all the failed assets the preview will now become available.
Note: When the user clicks “Generate video”, that is when the individual scene/video assets (image/video/voiceover) are all combined and rendered as a single video file.
Data Security Overview
Is the AI technology developed in-house or supplied by a third party?
AI Video is integrated into the SimGate platform via Third-party API from Pictory
Pictory Privacy Policy https://pictory.ai/privacy-policy
Does the client retain ownership of the data that it uploads?
Yes.
Anonymization - is data anonymized or aggregated in such a way that cannot be traced back to any specific individual or entity?
Yes, all data sent to-and-from AI services are subject to enterprise standards of data protection (e.g. end-to-end encryption, TLS, etc.)
Training - is customer data or content used in any way to train or improve AI services?
Pictory does not use client-uploaded or generated content for AI training purposes without explicit consent.
