The Corpus of German Speech (CoGS) is a 51-million-word corpus of geolocated automatic speech recognition (ASR) YouTube transcripts from local government channels in Germany, created for the study of lexical, grammatical, and discourse-pragmatic phenomena of spoken language, as well as for content and language analysis in digital humanities and social science fields. Annotation includes individual word timings and video IDs of transcripts, making it easy to instantly view the video(s) for any given search. The corpus was created from 39,495 ASR transcripts from 1,313 YouTube channels, corresponding to more than 7,223 hours of video. The size of the corpus is 50,514,575 tokens. The channels sampled in the corpus are associated with local government entities, mostly city governments. Related resources are the Corpus of North American Spoken English, the Corpus of British Isles Spoken English, and the Corpus of Australian and New Zealand Spoken English. The untagged corpus is contained in the file cogs_tokens_distributable.csv.gz. The PoS-tagged corpus with word timings is cogs_textpos_distributable.csv.gz. The separator for both files is the pipe character "|". The files have the columns 'Bundesland', 'Gemeinde', 'channel_title', 'channel_url', 'video_title', 'video_id', 'upload_date', 'video_length', 'location','nr_words','text_pos' (or 'text', in the untagged version), and 'latlong'. Each row corresponds to an individual transcript. In order to comply with Fair Use provisions of US copyright law, the original ASR transcript files from YouTube have been transformed in this version of the corpus: Every 200 words, ten words have been replaced with "@". In the PoS-tagged and word-timed corpus, these tokens have the form @_XX_1.0. See also https://cc.oulu.fi/~scoats/CoGS.html.

2022