What is it about?
Podcasts are a forma of media that has seen a massive increase in consumption in the last years. A podcast can vary a great deal in format at topics, with the majority of the shows dealing with multiple topics. In our work we experimented with automatically segmenting podcast shows from two new datasets into their topical components. We used recent advances both from the text and audio domain, providing the first work in our knowledge combining the two modalities within our framework.
Featured Image
Photo by Icons8 Team on Unsplash
Read the Original
This page is a summary of: Multimodal Topic Segmentation of Podcast Shows with Pre-trained Neural Encoders, June 2023, ACM (Association for Computing Machinery),
DOI: 10.1145/3591106.3592270.
You can read the full text:
Contributors
The following have contributed to this page
