Open Audio Search is a search engine for audio. It uses automated speech recognition to turn spoken word into text, which is then indexed in a full-text search engine. The engine can subscribe to RSS feeds to ingest content from different sources. Through a web user interface, users can search in the generated transcripts of radio shows and podcasts and play them right from the time mark of a search result. Open Audio Search is open source, developed as a collaborative project and can be easily installed on any server.
Open Audio Search is still in alpha and not yet stable for production. Feel invited to report bugs and tell us about ideas and missing features on our GitHub issue tracker or contact us by email if you're interested in using or supporting the project.


Open Audio Search is an open source project by arso collective and cba - cultural broadcasting archive.

Read about how to run your own instance, the technical details, and how to contribute to the project in our README


The development of Open Audio Search is supported by: