Audioburst is the largest audio library of spoken-word audio.
We listen and index 6 million minutes of broadcasted radio and podcasts every month(!)
The first thing we do with this content is transcribe it and extract audio cues (speaker change, silence, music, hesitations, background sounds, laughter, applause and more) in real-time, in order to get the most accurate topical segmentation.
Afterward, we analyze it and extract insights such as keywords, entities, sentiment, and additional metadata features.
The audio data is then organized according to the topic and stored in our repository for search and listening experiences.
Bursts – are short-form audio files that are created on-demand around a topic, context, theme, and genre requested by the user.
Our NLU (Natural-language understanding) makes each audio-clip searchable using natural language.
Bursts are an average of ~30 seconds to 4-5 minutes long and include content from over 100 different categories.
Pretty cool, right? 😊
Check out this great article to learn more about Audioburst Creators, and start creating bursts of your podcast episodes : What is Audioburst Creators?