This meeting focused on accessibility challenges in the streaming industry, with John Sweet from Paramount Streaming and Chris from AWS discussing captioning, subtitles, and sign language accessibility solutions. Chris shared his experience working with customers on live streaming caption requirements, highlighting the difficulties with multi-language support and the transition from hardware encoders to AI-powered solutions. They discussed technical challenges including the limitations of current protocols like 608-708 embedded captions and the need for better integration between live production and delivery systems. The conversation covered various accessibility needs including captions for the 3% of people who are deaf or hard of hearing, sign language support for the deaf community, and descriptive audio for the blind community. They explored different approaches to implementing sign language interpretation, including separate video assets and picture-in-picture displays, while acknowledging current limitations in AI-generated sign language avatars due to the “uncanny valley” issue. The discussion concluded with agreement on the need for industry-wide standards and coalition between major streaming companies to push vendors toward supporting multi-language captions and better accessibility features across the entire end-to-end workflow.
Chris
- Create a one-pager proposal on the recommended approach for multi-language caption/subtitle workflow and protocol, to be reviewed in the next meeting.
John
- Share the one-pager proposal with the group of accessibility leaders from different streaming and cable companies and arrange a call to discuss the proposal if there is interest.
- Connect with Chris on LinkedIn.
- Antitrust Statement & Accessibility in Streaming Discussion
- John led a discussion on accessibility in streaming, explaining the importance of addressing accessibility challenges and sharing statistics on disability demographics.
- He clarified that the goal of the working group is to discuss problems and find solutions without targeting specific companies or IPs.
- Chris, a senior solution architect at AWS focusing on media services, shared his experience working with customers on captioning and subtitles, noting the growing trend of younger generations using subtitles for content consumption.
- Both participants highlighted the importance of accessibility for both disability needs and audience engagement in streaming.
Multi-Language Captioning Streaming Challenges
- Chris discussed the challenges in enabling closed captions and subtitles for live streaming beyond English, particularly highlighting the difficulty in supporting multiple languages and the limitations of current hardware encoder workflows.
- He explained that while streaming delivery protocols like WebVTT support multiple languages, live production and media processing still face significant barriers, including the lack of proper multi-language caption protocols and limited AI integration in media processing units.
- Chris also touched on additional accessibility challenges including descriptive audio and sign language support, noting that these are often driven by large enterprises with specific accessibility strategies.
Sign Language Video Implementation Discussion
- Chris and John discussed the implementation of sign language interpretation and subtitles in video content, focusing on user experience and technical architecture.
- They explored approaches to displaying captions and sign language content, including the challenges of positioning, speaker identification, and user preferences.
- Chris emphasized the importance of making the process user-friendly by allowing customization on the client side while maintaining compliance requirements on the server side.
- They also discussed the potential for AI-generated sign language avatars and the need for seamless integration into live video workflows.
Sign Language Support Implementation Discussion
- Chris and John discussed implementing sign language support through avatars versus human interpreters for both live events and VOD content.
- John noted that while avatars could be useful for live events due to automatic generation, there are current limitations with facial expressions in sign language avatars, leading to a preference in the deaf community for human interpreters in VOD content when possible.
- They explored technical implementation options including overlaying avatars on the main video and delivering separate feeds side by side or top-down, with Chris suggesting that transcoding costs could be incurred upfront for VOD content to enable flexible delivery options.
Streaming Accessibility Features Discussion
- Chris and John discussed challenges and potential solutions for improving accessibility features in streaming, focusing on captioning, subtitles, sign language, and audio description.
- They identified key barriers, including limitations in current encoding protocols and the need for flexible, text-based subtitle tracks that can be easily embedded and remuxed for delivery.
- Both agreed on the importance of industry-wide collaboration to establish a common standard and encourage vendors to invest in new technologies.
- Chris offered to create a one-pager proposal for review, which John will share with a group of accessibility leaders from various companies.
- They plan to follow up in their next bi-weekly meeting.
Next Meeting: Wednesday, July 15, 2026 9 am PDT