Reading between the frames
An experimental data visualization project to manipulate youtube video and extract its latent information using IBM Watson's Natural language understanding tool.
Using IBM Watson's speech to text API, a transcript text is generated from the input youtube video link. That transcript is used to generate a word cloud using R. This word cloud depicts the highly frequent words used in the video, it's a quick way to skim the content of the video. Simultaneously this text transcript is also passed on to Watson's Natural Language Understanding API for analysis.
The major concept the of the video is identified. This will give the user a better classification of youtube video based on the concepts of the video.
Workflow of Reading between frames
