Linguistic Data Consortium resources

The Carnegie Mellon Speech Group is a charter member of the LDC and we have maintained an uininterrupted membership since the beginning. As a result we have (almost) all corpora released by the Consortium. The exceptions are some of the very early corpora (when LDC required purchase of individual corpora) and stuff that people took but never returned to the group's archive (don't let this be you!).

How do I find something?

First peruse the LDC catalog to locate what you need. You can then check our holdings for LDC has already sent us.

If you still can't find a copy of an LDC corpus, please get in touch. Be sure to have an exact catalog number for what you need. Many corpora have similar names or have colloquial names that only vaguely correspond to their actual names. It's your responsibility to resolve the reference.

Other corpora

Some people are under the impression that all corpora come from LDC. This is not true. Other sources include: ELRA/ELDA and NIST. Many of these corpora are kept in the library but are not catalogued on-line. You're welcome to look through the collection. Bear in mind that most of the NIST corpora were subsequently released through the LDC; you should look there first. We can offer advice on locating non-LDC corpora but please understand that doing so is primarily your reponsibility.

Note that Carnegie Mellon has an ELRA/ELDA membership which entitles us to get corpora at a discount. Please work through us to make sure that you get the discount.

Locations