Welcome to the MATCH Corpus

The MATCH Corpus is an output from the MATCH project which completed in October 2013. More details are available from the project website. To access the data, follow the directions given here.

This release of the corpus contains data from 22 younger and 24 older participants. Four of the original participants did not give their consent to us sharing their data with other researchers. Their data is used in our own research, but is not included in this distribution.

All of the audio, transcription and annotations have been released publicly under the MATCH Meeting Corpus license, very similar to the Creative Commons Attribution NonCommercial ShareAlike 2.5 License.

If you have issues with the MATCH corpus download please contact jonathan at inf.ed.ac.uk