Next: Data Protection
Up: Legal Aspects, Contracts
Previous: Client and Contractor
Contents
Once your speech corpus has been finished and you have checked out the
possibilities of distributing it to others, the next legal relationship
to consider is that of the copyright holder(s) and the user(s) of the
resource. As with software the copyright holders of language resources
want to protect themselves against unauthorized copying. There are no
technical ways to protect a speech corpus against unauthorized copying;
it would hinder the usage of the data too much. On the other hand it
is almost impossible to sue a user for copying speech data: in most cases
your `customers' will not even be in your country and proof of abuse will be
hard to establish. The best thing you can do is that
- you let the users sign a license agreement that clearly states the
rights of usage and clearly forfeits any rights to copy or re-distribute
the data, or
- you insert the conditions of usage in your corpus documentation and say
that the user, by using the speech data, automatically accepts the
conditions stated there.
Both methods are probably legally unsafe. However, the user of a speech
corpus is in most cases not a private person but either a company or a
scientific institution. It is not very likely that these types of
`customers' will explicitly commit fraud by re-distributing your speech
corpus.
Next: Data Protection
Up: Legal Aspects, Contracts
Previous: Client and Contractor
Contents
BITS Projekt-Account
2004-06-01