Scheduled at 11:30 in Denver 1-3 on Thursday, November 19.
#43203Speaker(s)
- Cindy Camp, Training Coordinator, DCMP
- Kyle Sisk, CTO, DCMP
Session Details
- Length of Session: 1-hr
- Format: Lecture
- Expertise Level: Beginner
- Type of session: General Conference
Summary
DCMP has developed a new video accessibility tool for adding captions, audio description, and sign language called “DCMP Create”. [KS2.1]DCMP Create is an entirely web-based platform that thoughtfully leverages AI to assist users, including those with little to no experience, throughout the process. This presentation will focus on DCMP Create and explain its benefits for the post-secondary community as they work to meet the new extended digital accessibility compliance deadline.
Abstract
The software and services for creating captioning, audio description, and ASL vary greatly in price, usability, and functionality. DCMP recognized the need for an intelligently guided DIY accessibility experience that could leverage its 30+ years of experience producing high-quality accessible media. Using machine learning, AI, and natural language processing, our team has created a web platform that enables users to submit videos and make them accessible. Additionally, users will be able to embed, download, or share links to their videos.
The captioning feature uses AI speech-to-text engines in tandem with NLP-driven timing and line-breaking guidance trained from our Captioning Key guidelines. This results in a highly accurate transcript that is properly segmented. A quick human quality control review is all that is needed to complete captioning for most videos.
Audio description is thoughtfully crafted narration that provides visual information in an audible form. Information that would be missed by blind or visually impaired viewers. The tool allows users to write audio description with the assistance of image description AI, record narration for those description segments, or select a neural synthetic voice and generate the narration. Segments may be in-line or expanded, and the tool will fully mix the program and audio description segments. Users may embed, share links, or export audio files for use on other platforms.
Synchronized ASL capture typically involves a camera and post-production in video editing software such as Adobe Premiere. DCMP’s tool allows users to record from their phone, tablet, or computer webcam. The original video is counted in and played alongside the ASL capture to reinforce synchronization with the original video. Multiple takes, edits, and alignment all happen within a simple web interface. Users may embed the ASL (picture-in-picture) or place it beside the original video and share via embed link or a video export.
Keypoints
- Current video accessibility apps have a high learning curve & are inaccessible to many disabled users.
- Adding captions and audio description to videos is not sufficient. They must be high-quality.
- DCMP’s Accessibility Tool can help colleges become compliant with new digital accessibility regulations.
Disability Areas
Cognitive/Learning, Deaf/Hard of Hearing, Vision
Topic Areas
Accessibility and Accommodation in the Current Political/Legal Environment, Accessible Course Design, Alternate Format, Artificial Intelligence, Assistive Technology, Campus & System-wide initiatives, Captioning/Transcription, Faculty Development & Support, Teaching about Accessibility in Curriculum, Uncategorized
Speaker Bio(s)
Cindy Camp
Cindy Camp is the Training Coordinator with The Described and Captioned Media Program. She holds a Master’s degree in English, is a nationally certified interpreter, certified in mental health interpreting, as well as a C-Print captioninst and trainer. She has provided access services for students with disabilities for over 30 years.
Kyle Sisk
Kyle Sisk is the Chief Technology Officer with the Described and Captioned Media Program. He has been with DCMP for 17 years.