Introduction
Your Shape platform helps you create learning assets (called Shapes) within minutes, that you can share with other users and with your team. Thanks to a number of customization options, you can create hyper-personalized content to best fit your needs.
This article will guide you through the creation of a new Shape from a document, link, or custom text. Alternatively, you can create a Shape from a storyboard, and choose by yourself the text that will appear in each scene, or create a Shape from scratch.
How to get the most out of Shape’s Artificial Intelligence.
Creating a New Shape from a Document, Link or Custom Text
The main feature of the Shape platform is creating content using Artificial Intelligence.
To create a new Shape, log into your platform, click on the New Shape button in the top right corner of the page, then select your source: document, link or custom text. Artificial Intelligence will analyze the content you insert and as a result it will create your new Shape.
Shape’s Artificial Intelligence works with content in the following languages: English, French, German, Italian, Portuguese, and Spanish. If you provide sources in other languages, the creation process will fail.
Also remember that the maximum length of the content you can upload is 10,000 words. This limitation applies to all of the types of source and file formats supported. If your content is longer than 10,000 words, you can split it into multiple files and then upload each file separately.
Document Source
If you want to start from a document, on the page that opens click on the Document item to upload a file from your device. You can also drag and drop the file anywhere on the page to upload it. Click on the Supported formats link to view the list of the supported file formats.
Refer to the Best Practices chapter to get the most out of document-source Shapes.
This is the list of supported file formats (here you will also find sample files structured so that Shape’s AI can easily process them):
File Format |
Notes |
|
The quality of the output could be low if the document has at least one of the following characteristics:
|
DOC, DOCX, ODT, RTF |
The quality of the output could be low if the document has at least one of the following characteristics:
|
PPTX |
The quality of the output is low if the presentation is made of multiple slides containing few words each |
TXT |
/ |
JPG, JPEG, PNG |
We perform an Optical Character Recognition process to extract text from the image. The text length needs to be at least 20 words to produce an acceptable output. |
HTML, HTM |
When choosing the type of source, select the Link source type and not the Document source |
Link Source
If you want to start from a URL, on the page that opens click on the Link item, then paste your link in the corresponding field and press Create Shape.
When using a link as your source, make sure that the webpage you are linking is public and does not require authentication to be accessed. The same applies to documents on Google Drive, OneDrive or other repositories — files need to be shared publicly to be used by Shape.
Refer to the Best Practices chapter to get the most out of link-source Shapes.
Free Text Source
If you want to start from some text, on the page that opens click on the Free Text item, then type or paste your text in the corresponding field. You can use the text editor to apply formatting options to your text. Remember to provide text longer than 50 words.
Refer to the Best Practices chapter to get the most out of free-text-source Shapes.
When ready, press Create Shape.
Shape’s Creation Process with Artificial Intelligence
Once you have chosen the type of source, you are ready to create your Shape. On the page that opens, you can follow the entire creation process, starting from the extraction of text up to the production of new content, your Shape! Let’s go through each one of the steps of the process:
- Extracting Text. Shape’s Artificial Intelligence starts by analyzing the source you uploaded and extracting texts from it
- Identifying Topics. AI identifies different topics within the text of the source, splitting them into blocks
- Extracting Concepts. Then, it summarizes each block just created by extracting the most important information
- Finding Media. When the text is ready, Shape’s AI searches for media assets that fit with the text of each block, and uses them as a background. Assets (images and videos) are free from copyright and are downloaded from Pexels (opens in a new tab).
- Producing Content. In the last step, Shape combines all of the information retrieved in the previous steps, producing a sequence of scenes resulting in a learning asset.
The creation process may need from 5 to 10 minutes to complete.
If you want to stop the creation process, you are free to do it at any time by pressing the Stop Creation Process button.
While waiting, you have the possibility to choose how long your content will be — from short content (small — about 10% of the source content length), to medium length (about 40% of the source content length), up to long content (large — about 70%). If you don’t select any options, medium size will be the default for your Shape. You can choose the size before the Extracting Concepts step, then this panel will disappear when the Extracting Concepts step starts.
After the Extracting Concepts step, you can also choose a title among three titles extracted by AI. You are free to write a new title if you want. If you don’t select any title, Shape will choose one as default, but you can change it later on.
You don't need to wait that the creation process is completed, you can close the creation panel using the x icon whenever you want. The process will continue in the background and you can check it from the homepage. When your Shape is ready, you will receive a notification. Press the bell icon in the bar at the top of the page and open the notification to view your new Shape.
You will also find the Shape just created in the All Shapes and in the Created by me tabs on the homepage, in Draft status. Now, you can preview, edit, and publish your content. Learn more about editing your content and publishing it.
Best Practices
The AI processes behind Shapes are trained to extract the text from the source, split it into blocks, and then summarize each block. Shape’s algorithms have been trained with a very large document set, and some conditions can help the text extraction.
To get the most out of Shape’s AI, make sure that you structure your document, webpage or free text as a Wikipedia page (opens in a new tab), using titles, subtitles and images. Find here an example of how to structure your source (PDF, 200 KB).
Watch the following video, created using Shape, for a summary of the best practices of use.