Highlights –

  • The Twelve Labs search technology lets viewers quickly jump to a video spot using the natural-language queries.
  • The startup will utilize the infrastructure resources of Oracle’s cloud platform as part of the agreement to support its AI development efforts.

Twelve Labs Inc. has secured USD 12 million in funding from investors to develop a search platform that allows viewers to find interesting parts of a video quickly.

The corporation has just announced the investment. The funding was given as an addition to a seed round that had already closed earlier in the year. Radical Ventures headed the most recent round of financing for Twelve Labs, with participation from Index Ventures, Spring Ventures, WndrCo, and many other angel investors.

Using natural-language queries, users of the search platform from Twelve Labs can swiftly jump to a particular spot in a video. The platform, for instance, might be used by a marketer to locate parts in TV commercials where a company’s logo can be seen. Users can also look for other items, audio clips and text excerpts like handwriting on a whiteboard.

Organizations can tailor the startup’s platform to meet their needs. Customers can use specific datasets to develop the artificial intelligence model that powers its search platform, in particular. Processing accuracy is increased when an AI model is trained on a dataset comparable to the information a business intends to evaluate in production.

Users can access the platform through a collection of cloud-based application programming interfaces. Twelve Labs claims that with just 20 lines of code, APIs can help developers to integrate its platform into an application. According to the startup, its early users have already used its APIs for various use cases, from online education to advertising.

According to Radical Ventures Partner Rob Toews, “Twelve Labs is opening up entirely new possibilities for the future of video. The team’s deep commitment to research and ongoing innovation is driving an entire industry forward to efficiently access and fully leverage 80% of the world’s data that currently resides in video form. Twelve Labs is positioned to build the world’s leading foundation model for videos, forever changing how we interact with this medium.”

The additional funding of USD 12 million that the business has obtained from investors will be used to expand the capabilities of its platform. The business claims that during the past six months, its developers have more than doubled the accuracy of its video search algorithms. In the future, it will expand the platform’s functionality and invest in AI research.

The startup intends to construct a new AI model optimized for video processing with several billion parameters as part of its research efforts. The configuration options that specify how a neural network processes data are known as parameters. A neural network can perform processing more accurately when the parameters are more.

Twelve Labs claims that a variety of jobs related to video processing will be handled by its upcoming AI model. According to the business, the AI model will be well-suited for use cases like automatically categorizing videos into chapters and producing clip summaries.

To create the AI model, the business has collaborated with Oracle Corp. As part of the collaboration, the startup will use Oracle’s cloud platform’s infrastructure resources to help its AI development activities. An infrastructure environment with approximately 1,600 gigabits per second of network bandwidth and hundreds of Nvidia Corp. graphics cards has been created by Twelve Labs.