FIAT/IFTA World Conference 2025

Slowly but steadily the podcast as a format gained more and more power as a dominant form of media. It is the way in which millions of people consume news, politics, entertainment and gossip on a daily basis. So since 2021 we’ve been actively working on preserving these audio stories that are created both by media professionals and hobbyists with a microphone on the kitchen table.

In this talk, we share key insights into how we’ve been archiving and preserving podcasts from the Netherlands for over four years. Why, after studying the distribution models of podcasts, we decided to ignore playback platforms like Apple Music or Spotify, but make use of a podcast RSS aggregator service instead. Using the Listennotes API, our script allows us to automatically gather podcasts in MP3 format together with any descriptive metadata that's included in the RSS feed by the podcaster creators. Simply adding new shows to a playlist enables us collect the latest episodes on a weekly basis. As we will walk you through our method, we go in-depth as to how we addepted MP3 as an accepted file format to ingest podcasts in our infrastructure, how we enrich episodes with additional metadata and make the shows accessible on our platforms to users. We explain our selection process using license agreements with creators and how we’re trying to get as wide of a vertical slice as possible of the Dutch podcasting landscape. Finally we address paywall related challenges that have become more frequent and that we are struggling with.This talk provides pointers that will allow anyone to get a grasp on how to preserve podcasts and make sure these stories can be told for generations to come.

The extensive audiovisual archives of Radio Televisión Española (RTVE) represent a rich, underutilized resource for production research. A particularly valuable yet challenging collection is the RTVE Film Archive, encompassing news programs from the 1950s to the late 1980s. While undergoing digitalization to improve accessibility, a significant portion of this news collection lacks accompanying sound and suffers from non-informative titles, hindering effective retrieval.

Our central research questions focus on how to effectively explore this news film collection for production needs, evaluate the performance of multimodal information retrieval technology in this context, and determine the feasibility and integration strategy of resulting metadata into the existing archive management system.

Focusing exclusively on non-textual, video-only content, we explore how to effectively represent video content in the absence of accompanying text or metadata by embedding both textual queries and video visual features into a shared vector space. This methodology utilizes the CLIP model to generate embeddings of video frames and natural language queries, constructing a multimodal semantic space where semantic similarity is measured by the proximity of representation vectors. The VTR system pipeline encompasses video analysis, semantic analysis, vector database ingestion, and search with relevance feedback, allowing users to query the film archive using natural language or example images.

The main contributions of this PoC include a demonstration of a functional VTR system for retrieving information from a film archive without relying on existing metadata, an evaluation of the effectiveness of joint semantic space representation for this task, and insights into the potential and challenges of integrating such technologies into existing media archive workflows.