What is qiaomu-youtube-ai-podcast?

This project is the source code and data behind a website that curates AI related podcasts and video interviews for a Chinese speaking audience. The live site organizes the shows by topic, such as research, engineering, startups, news, business and culture, so someone interested in AI does not have to hunt across many scattered sources to find good listening material. For each episode, the project adds a Chinese title and a short Chinese summary alongside the original English version, since scanning long English titles and descriptions can be slow for readers who prefer Chinese. It also tracks, for every episode, whether a full transcript exists, whether there is a deeper AI generated summary based on that transcript, or only a fallback summary based on the description alone. This lets a reader judge quickly whether an episode is worth their full attention. At the time of the last recorded check, the index covered twenty five podcasts, one hundred twenty five episodes, ninety nine transcripts, and one hundred twenty five summaries in total. The whole thing builds into a plain static website with no server side application needed, so it can be hosted on any basic web server or content delivery network once built. To run it yourself you install the dependencies, run a data validation check, build the site, and then serve it locally to preview in a browser. There are also optional scripts for fetching new episodes, translating titles, importing captions, and generating summaries, though several of those rely on the original author's personal accounts and are not fully usable out of the box. The repository intentionally excludes any API keys, account credentials or private logs, and the code itself is released under the MIT license, while the podcast audio and show notes remain the property of their original creators.