A speech-to-data platform for real-time media analysis (100% in ideation).
- Click into a stream, say, a presidential debate, viewer sees a real-time speech-based analysis of how the topic is being covered. A little bit of visual inspo
- In a big, media-wide topic like the Brett Kavanaugh hearing, the platform could show a real-time speech-based analysis of how digital news media covers the topic.
- Long term, platform could watch and store all streamed data to allow for real research into how digital news media covers different topics. Inspo: Media Cloud.
- Layout inspiration: The Times' real-time analysis of the 2016 debate
- What if the transcript were layed over the video, and then the journalists who were live reporting this could highlight/comment on a piece of text? The text would be highlighted like in @brtacks/foundations and the annotation would appear on the side.
- Home page sort of looks like the home page for a [insert streaming service here] app on Apple TV, where there are clips and previews of all the stuff that's streaming right now (Dana Bash on CNN, Anderson Cooper, Tucker Carlson on Fox News, a football game/pregame/postmatch).
- Watson websocket
- Lots of D3