OCI Speech provides automatic transcription and captioning for customers of the Kaltura video platform

October 18, 2022 | 2 minute read
Jay Hardikar
Senior Principal Product Manager
Text Size 100%:

Kaltura is a leading provider of live, real-time, and on-demand video software-as-a-service (SaaS) solutions across many industries. It provides video solutions to thousands of organizations around the world, engaging many millions of viewers at home, at work, and at school. It provides an array of services for enterprises, educational institutions, media companies, and developers.

Kaltura’s platform, solutions, and services help their customers with everything from live video events and video messaging to managing media assets and multimedia classrooms.

The problem

Customers of Kaltura who want to perform tasks such as captioning, transcription, translation, or in-video search look to Kaltura’s video enrichment marketplace called REACH. REACH provides a set of automated and human services to Kaltura’s customers through a network of partners.

Kaltura was looking to provide an accurate, fast, and cost-effective automatic transcription service as part of its REACH marketplace to offer to their customers.

The solution: OCI Speech service

Oracle Cloud Infrastructure (OCI) Speech is a service that uses automatic speech recognition (ASR) technology to transform audio-based content to text using prebuilt acoustic and language deep learning models. With features such as native multiple language support, integrated transcription service, and profanity filtering, it allows for very high-speed transcription and captioning of audio and video content. OCI Speech models use enterprise domain knowledge resulting with high accuracy transcriptions. The Speech service protects the customers’ privacy by not storing any customer content.

Applying the rich API of OCI Speech and Kaltura’s platform, Oracle and Kaltura jointly developed a solution for the REACH marketplace that is seamless and transparent to Kaltura’s customers, while taking advantage of all the benefits that OCI Speech offers.

The benefits

OCI Speech’s automatic transcription and captioning on Kaltura’s REACH platform provides the following unique benefits to Kaltura and its customers:

  • Lower cost automatic captioning service than current alternatives: The Speech service runs on OCI, a cloud platform built for enterprises seeking higher performance and consistently lower costs.

  • Confidence score for each transcribed word: Customers can identify words that have been transcribed correctly with high confidence or not, allowing visibility into which words need manual checking and correction.

  • High speed transcription and captioning: OCI Speech uses smart splitting and stitching algorithms, which allow for transcription processing of hundreds of segments of audio in parallel. This functionality provides customers with high-speed automatic transcription and captioning.

How to use OCI Speech for your transcription needs

With Kaltura, customers can use OCI Speech to automatically add subtitles to their audio and video files to reduce the production time and make it easy to publish their content. Kaltura REACH integrates seamlessly on the backend with OCI Speech service. Kaltura customers don’t need to create any separate account on OCI or perform any tasks outside their regular REACH workflow. Customers of Kaltura can use this service simply by uploading their files to Kaltura REACH, which automatically transcribes the files using the OCI Speech service.

For all your video transcription needs using Oracle Cloud Infrastructure Speech, check out Kaltura’s REACH marketplace.

Jay Hardikar

Senior Principal Product Manager


Previous Post

Industrialization of artificial intelligence and machine learning for SaaS applications

Miroslav Novak | 17 min read

Next Post


Will increasing regulation stop businesses using AI?

John Menhinick | 4 min read