20230129431. One-to-Many Automatic Content Generation simplified abstract (salesforce.com, inc.)
Contents
One-to-Many Automatic Content Generation
Organization Name
Inventor(s)
Michael Sollami of Cambridge MA (US)
Sönke Rohde of San Francisco CA (US)
Alan Martin Ross of San Francisco CA (US)
David James Woodward of Bozeman MT (US)
Jessica Lundin of Seattle WA (US)
Owen Winne Schoppe of Orinda CA (US)
Brian J. Lonsdorf of Soquel CA (US)
Aashish Jain of Cambridge MA (US)
One-to-Many Automatic Content Generation - A simplified explanation of the abstract
This abstract first appeared for US patent application 20230129431 titled 'One-to-Many Automatic Content Generation
Simplified Explanation
The patent application describes a technique for automatically generating new content using a trained generative adversarial network (GAN) model. This technique can save time and computing resources in the content generation process.
- A computer system receives a request for newly-generated content, including the current content.
- The computer system uses a trained 1-to-n GAN model to automatically generate n different versions of new content.
- Each version of new content is generated based on the current content and one of n different style codes.
- The computer system then transmits the n different versions of new content to a computing device.
Potential applications of this technology:
- Content generation in various fields such as art, design, writing, or marketing.
- Personalization of content based on different styles or preferences.
- Automated creation of variations for A/B testing in marketing campaigns.
Problems solved by this technology:
- Manual content generation can be time-consuming and resource-intensive.
- Generating multiple versions of content with different styles can be challenging for human creators.
- This technique automates the content generation process, saving time and resources.
Benefits of this technology:
- Saves time and computing resources by automating the content generation process.
- Enables the generation of multiple versions of content with different styles.
- Provides a more efficient way to personalize and test content.
Original Abstract Submitted
techniques are disclosed for automatically generating new content using a trained 1-to-n generative adversarial network (gan) model. in disclosed techniques, a computer system receives, from a computing device, a request for newly-generated content, where the request includes current content. the computer system automatically generates, using the trained 1-to-n gan model, n different versions of new content, where a given version of new content is automatically generated based on the current content and one of n different style codes, where the value of n is at least two. after generating the n different versions of new content, the computer system transmits them to the computing device. the disclosed techniques may advantageously automate a content generation process, thereby saving time and computing resources via execution of the 1-to-n gan machine learning model.
- Salesforce.com, inc.
- Michael Sollami of Cambridge MA (US)
- Sönke Rohde of San Francisco CA (US)
- Alan Martin Ross of San Francisco CA (US)
- David James Woodward of Bozeman MT (US)
- Jessica Lundin of Seattle WA (US)
- Owen Winne Schoppe of Orinda CA (US)
- Brian J. Lonsdorf of Soquel CA (US)
- Aashish Jain of Cambridge MA (US)
- G06F3/04845
- G06N3/04
- G06V10/771
- G06V10/762
- G06V10/82