Dear Deva,
Thank you for the additional information.
You are right, I'd better clarify what I'm trying to do. If, in light of these clarifications, other works come to mind, please do share them.
My goal is to collect a fairly large number of Sanskrit texts, digitise the ones that haven't been digitised already, but mostly to find well-executed translations of the large volume of already digitised Sanskrit works. The user would then view the translation side by side with the Sanskrit original, which has a number of linguistic benefits compared to lookups of single words (although that functionality will also be available).
Such a corpus can be used in many ways.
For example, if several translations of the same Sanskrit text are found (into different Indian or non-Indian languages), there arises the opportunity to gradually get to know these languages. Today Indians are encouraged to learn other Indian languages, which is good advice, but assembling a corpus as described above can make the acquisition of other Indian languages easier for people who already know one of them.
Or consider the example of a scholar working on a new translation of a Sanskrit text, or perhaps wanting to read a given text with deeper understanding. Or maybe a student learning Sanskrit. He/she could search for the occurrence of a given phrase within the Sanskrit corpus, and then, depending on which languages are known to this person, see how this phrase has been translated by others, as well as the contexts in which the phrase occurs in other Sanskrit texts. Obviously, the more languages the researcher knows, the more translations he/she would be able to make use of and the richer the reward from using the corpus. And if some translations happen to be fanciful, then looking at other translations might mitigate this. There will be a way for the user to customise their workspace by defining which texts and which translations they prefer. Or, if they want to see everything, then they can do that too. On the more technical side, the further automated Sanskrit corpus annotation develops, the more sophisticated tools will be at the disposal of technically savvy users. But there is plenty for everyone even without that functionality.
While the uses of such a parallel corpus can be manifold, but I make no secret of the fact that my key goal is to advance Sanskrit knowledge worldwide by creating a flexible platform to which texts, translations and dictionaries could be added. I also think that Sanskrit studies would benefit from bringing into the picture other Indian languages, since the many linguistic commonalities there can't be ignored.
Does this explanation help? If not, please reply with questions, and I'll clarify further.
Irene