This week we report on new ideas for ±«Óãtv Box, Audio AR and the new W3C Datacue API
±«Óãtv Box Ideation
The Better Internet team have run a series of ideation sessions around the ±«Óãtv Box project. The session began with generating many ideas quickly which were then categorised into broader themes. The team then took 10 ideas forward for further development and made a catalogue of them to use as discussion points with the Box team.
Some of the ideas discussed were a 'Scrapbooking Box - a clever way to archive your social media data and share memories with friends.', 'Mood based reminiscing - a smart printer that cheers you up with little printed reminders of happy days' and my personal favourite the 'Robotic Self Oracle - a chat bot that gives you answers based on your previously shared tweets and slack comments'.
Managing datasets
The Data team have been working on the bbc-data tool for distributing and managing large machine learning datasets and models. They have finished a large refactor of the codebase and implemented the to ensure data validity and to store package metadata. The Bagit spec is widely used in digital libraries and a lot of the code has been written by the .
The team have also been refactored the code they use to interact with Amazon S3, where that datasets are stored to be shared. The tool can now very quickly write a compressed archive directly to the S3 service using many parallel HTTP requests. This is particularly useful when uploading very large datasets of 10s or 100s of gigabytes.
W3C Data Cue API
Chris has been participating in W3C discussions around the new , which is an API for aligning metadata against AV material. This is really interesting to our work here in R&D with lots of potential for triggering various events in Object Based media experiences.
The group have been looking at how the API can be built on top of existing browser support for subtitles and captions and are assessing the consistency of current implementations. The next steps are to liaise with the browser vendor development teams to get consensus and being on the specification.
Audio AR - Smart Headphones
The collaborative project with Anna and Valentin from QMUL has been progressing with two successfully pilot tests of the 'I am not a robot' experience. Meanwhile the team here in ±«Óãtv R&D has recorded version one of their own prototype 'Looking for Nigel'.
Holly has been researching methods for measuring and evaluating audio AR experiences with the aim of constructing our own bespoke methods. The development team attended the Unity Developer Day for an insight into future Unity techniques for rapid prototyping and emerging tools such as
-
Internet Research and Future Services section
The Internet Research and Future Services section is an interdisciplinary team of researchers, technologists, designers, and data scientists who carry out original research to solve problems for the ±«Óãtv. Our work focuses on the intersection of audience needs and public service values, with digital media and machine learning. We develop research insights, prototypes and systems using experimental approaches and emerging technologies.