https://cuddly.tube/videos/watch/1dca3cb9-6dc7-4958-a26e-459b38fde6fe None Video content is everywhere, but extracting key information from hours of recordings is time-consuming. Cloud-based transcription and summarization services exist, but they come with privacy concerns, recurring costs, and dependency on external providers. What if you could run the entire pipeline on your own infrastructure? This talk demonstrates a complete, self-hosted workflow for transforming video content into text summaries using only open-source tools. Starting with video files, we extract audio, transcribe it using OpenAI's Whisper, and generate concise summaries with open large language models—all running on-premises without sending your data to third-party services. This is about digital sovereignty and practical AI deployment. Whether you're concerned about data privacy, want to avoid cloud costs, or simply prefer to control your own infrastructure, this talk shows you how to build a production-ready video summarization system. Licensed to the public under https://creativecommons.org/licenses/by/4.0/ Tue, 14 Apr 2026 09:31:01 GMT https://validator.w3.org/feed/docs/rss2.html PeerTube - https://cuddly.tube https://cuddly.tube/lazy-static/avatars/81bc0609-c67e-4876-92ac-ee9c14bb4d24.png https://cuddly.tube/videos/watch/1dca3cb9-6dc7-4958-a26e-459b38fde6fe All rights reserved, unless otherwise specified in the terms specified at https://cuddly.tube/about and potential licenses granted by each content's rightholder.