Scheduled at 10:30am in Penrose 2 on Friday, November 10.#38158
- Vijayshree Vethantham, Senior Vice-President, Growth & Strategy, Continual Engine US LLC
- Length of Session: 1-hr
- Format: Lecture
- Expertise Level: All Levels
- Type of session: General Conference
Video accessibility is a moving target for most institutions and organizations. While automation has made an impact in making videos accessible, the secret lies in artificial intelligence (AI) combined with human review. We share our experience in utilizing advanced AI technologies and collaborative approaches to accelerate the audio description process, by automatically generating alt text for contextual and complex images and diagrams in videos.
Generative Artificial Intelligence (AI) is fast emerging as a game-changer in the field of education. Generative AI-based transcription automation reduces the time and cost of manual transcription, enabling educators to create inclusive content for all learners, including those with hearing impairments or speaking different languages.
Our presentation focuses on the successful implementation of AI technology in developing a proprietary caption management framework. Utilizing cutting-edge video processing modules and custom natural language processing (NLP) routines, Continual Engine (CE) has achieved contextual precision. We will include case studies of 2 higher education clients - CE fast-tracked their accessibility journey by processing various types of videos efficiently, including lecture recordings, STEM content, and video class recordings with unclear audio, within the committed timeline and at a cost competitive with industry standards.
We will discuss the key factors that have contributed to the widespread adoption of these technological innovations, ensuring that media content is accessible to all individuals, regardless of their abilities. Continual Engine’s proprietary technology, Invicta™ creates accurate and accessible closed captions and transcriptions of various educational content, including lectures, presentations, and videos. CE is expanding its video processing framework to become multilingual, providing the capability to handle and process content in a variety of languages. The automation process includes human experts for content verification. The market-tested approach ensures efficiency and quality control, resulting in a fast and cost-effective process that saves time and costs while maintaining quality output. The caption management framework generates outputs in multiple formats, including open and closed captions, transcripts, and audio descriptions. This flexibility ensures users can choose the format that suits their specific needs.
- Understand how to leverage generative AI for extended audio descriptions and closed caption
- Explore how well-designed AI can substantially enhance precision and minimize inaccuracies
- Learn about the potential of generative AI in closed captions, to assist in identifying and correcting errors
All Areas, Cognitive/Learning, Deaf/Hard of Hearing, Vision
Accessible Educational Materials, Assistive Technology, Captioning/Transcription, Uncategorized, Web/Media Access
Vijayshree has nearly two decades of experience leading multidisciplinary teams, and managing key client partnerships in higher education and accessibility as part of the founding team of two education-based start-ups – ansrsource and Continual Engine. Her experience includes building impactful partnerships with large publishers and institutions, understanding their content and learning goals, and guiding solutions to enable scalable, and accessible learning experiences. Vijayshree leverages her deep knowledge of start-ups, higher education, and custom content development along with the potential of AI and technology in education to create thriving engagements with educational providers, learning organizations and other partners. Over the last few years, Vijayshree has dedicated her time to exploring how robust and pragmatic educational technology, designed with the intention of solving a problem, can enable transformation, inclusion, diversity, accessibility, and affordability for everyone.