Godard’s recovered history of cinema: A case study in preservation, access and AI-based restoration

Abstract

This paper examines an exploratory use of artificial intelligence and natural language processing tools to restore and re-version audiovisual assets held by the Visual Collections Repository of Concordia University. The case study on which this paper is based focuses on magnetic tape recordings of a series of lectures delivered by Swiss-French New Wave filmmaker Jean-Luc Godard to cinema and humanities students in the late 1970s. The tapes are especially valuable to biographers of Godard and researchers studying the French New Wave but have never been exhibited in their entirety as the poor audio quality has historically limited their exhibition and research value. This case study describes innovative workflows for incorporating AI into restoration projects, particularly those involving audio reformatting. For audiovisual archivists, this has relevance to collections of deteriorated quality. The paper provides lessons and best practices when working with WhisperAI, Adobe Audition and iZoTopeRX software, in addition to a discussion of the broader questions surrounding AI’s role in cataloguing and creating access copies of archival master files. This article is also included in The Business & Management Collection which can be accessed at http://hstalks/business.