Datasets

Modern Times 1936 will use two datasets from the twentieth century: 80,000 digitised photographs from the 1930s available via DigitaltMuseum, and 900 hours of video from all weekly newsreels and short films produced by Svensk Filmindustri from the 1910s through the 1960s. Since most of this content belongs to the public domain the datasets will be curated and made publicly available, uploaded online at for example Github. More info at https://github.com/modern36.

Journal Digital Corpus

Speech recognition and transcription tool

Intertitle detection and OCR tool