Inspiration

What it does

Analysis of commentator audio to find highlights in a game

How we built it

We use the Symbl.ai's API to convert speech to text of a sports event's audio, and process that text.

Then we use NLP to find the tone of a certain phrase and find the respective phrase's timestamp.

For finding the highlight moments we focus on matching phrases with a "surprise" emotion factor in them.

Challenges we ran into

Getting the timestamp from the API response. We finally accomplished this by using the beta version of the API.

Accomplishments that we're proud of

What we learned

What's next for Sports highlight generation

Built With

Share this project:

Updates