The first public corpus of alcoholized German speech

Collectors of alcoholized German speech, rejoice! A new corpus is available to you:

Alcohol Language Corpus. The first public corpus of alcoholized German speech,” Florian Schiel [pictured here], Christian Heinrich, and Sabine Barfüßer, Language Resources and Evaluation, 2011. The authors, at the Bavarian Archive for Speech Signals, Ludwig-Maximilians-Universität München, explain:

Schiel“The Alcohol Language Corpus (ALC) is the first publicly available speech corpus comprising intoxicated and sober speech of 162 female and male German speakers. Recordings are done in the automotive environment to allow for the development of automatic alcohol detection and to ensure a consistent acoustic environment for the alcoholized and the sober recording. The recorded speech covers a variety of contents and speech styles. Breath and blood alcohol concentration measurements are provided for all speakers…. ALC is available without restriction for scientific or commercial use at the Bavarian Archive for Speech Signals….

The number of word tokens in the sub-group spontaneous speech reveals that in average speakers utter more words per recording item in sober condition than being intoxicated… The reverse effect can be observed in the sub-category command & control, in which the speakers were asked to read or formulate commands to the automobile… this is not what we expected to see, since the higher mental load required to think of new commands was expected to be diminished under the influence of alcohol and hence the number of words to be less than in the sober state.”


(Thanks to investigator Bette Dalkey for bringing this to our attention.)

BONUS: Baumeister B, Schiel F (2010): On the Effect of Alcoholisation on Fundamental Frequency. IAFPA Annual Conference 2010, Trier, Germany.