What Software Tools are Needed for Oral History?

Oral history is a powerful means of preserving and sharing the stories and experiences of individuals and communities. As memories fade and generations pass, the importance of documenting these narratives becomes evident. Setting up an oral history archive requires careful planning and the right tools to ensure that these valuable accounts are preserved for posterity. 

However, many projects with very valuable material are often inaccessible for the public. Videos are time-consuming to catalog and challenging to use for research if not searchable in full-text or through rich metadata. This post gives an overview of a series of tools that can help make an archive accessible and open. 

Transcription Software

Recording interviews is only the first step; creating accessible and searchable transcripts is equally vital. Transcription software can significantly streamline this process. By automating the initial transcription, you can save time and resources. However, manual review and editing are mostly still essential to ensure accuracy.

Archival Storage Solutions

Preserving the collected interviews in a safe and organized manner is critical for the long-term success of your oral history archive. Acquiring robust archival storage solutions, such as external hard drives or cloud storage services, to securely store and protect the digitized interviews is of key importance. Implement a consistent naming and cataloging system to maintain order and facilitate future retrieval.

Online Catalogue

Making the oral history collections accessible through an online platform is the best way to guarantee a broad access. In order to offer an efficient search experience, interviews should be accompanied by comprehensive metadata with descriptive information such as interviewee names, dates, locations, as well as keywords referring to the topics mentioned in the conversation.

AI-powered Indexing Assistance

Creating rich metadata for an oral history archive might be too time consuming and therefore impossible to achieve with available resources. Luckily, artificial intelligence can help. With natural language processing methods such as computing semantic similarity between transcript fragments and a given vocabulary or formulating descriptive keywords with generative language models can help save time. Through semi-automated systems or human-in-the-loop approaches (for example enabling the curation of the generated terms before assigning them automatically), archivists and researchers should be able to gain efficiency without compromising the control over relevance and accuracy.

Semantic Search Engine

Searching for specific topics might be especially challenging in oral history collections. It is likely that researchers search for abstract terms corresponding to academic literature, or journalists look for terms used in media, but none of this match the way people talk in everyday language. To avoid a frustrating search experience, integrating an AI-powered semantic search engine might be the best solution. By understanding the intent of the search and understanding the meaning of the words contained in the interviews, semantic search is capable of finding relevant results even without an exact match of keywords by computing semantic similarity.

aureka for Oral History

aureka is a web application especially designed to catalog and research with audio and video materials. It is ideally suited for oral history archives, assisting throughout the whole process of organizing, analyzing and publishing your collections.

• aureka can automatically transcribe the recordings and, if desired, transcripts can be manually edited to ensure that they are 100% correct
• As a cloud based application, it can store large amounts of videos and audio and make them available anywhere
• With aureka is possible to explore oral history collections in innovative ways thanks to interactive visualizations portraying the names of people, places and organisations mentioned in the recordings
• You can also add annotations synchronized with the transcripts
• There is a dedicated area to add metadata to each recording
• Transcript, annotations and metadata are fully searchable. Thanks to synchronization between text and audio it is possible to navigate to the exact point in which the searched word was said
• We will soon launch a publishing function to make collections accessible to external users directly through aureka as well as an AI-powered automated indexing function to save time in the cataloging work