http://opus.lingfil.uu.se/ is a nice resource of parallel corpora for many language pairs.

You have 13.4M tokens for (en->hi).  Not much for Marathi though..(en->mr 6.2M)

Thanks a lot! I had a look at the link. Looks like pretty comprehensive sets. However, after selecting En-US->Hi OR En-US->Mr, I found only tokens in KBs.

Am I searching it right? 

Edit - Oops sorry, I selected En rather than En-US and then it listed better list. Thanks for your help!

Dear madam

If you elaborate more on your parallel corpora requirement then I will be able to suggest you appropriate resources i.e what sort of parallel corpora ( parallel words or parallel sentence), domain of the corpora such as tourism, healthcare etc. One of the resource available for parallel corpora for Indian languages that includes Marathi also is http://www.tdil-dc.in. You have to register on that portal and send a hard copy of filled form to their office located in Delhi.

If you provide more elaborate information, then I will be help you in a better fashion.

Fill free to communicate your any query

Hi Goraksh,

Thanks a lot for your inputs!

I need parallel sentence based corpora. I will definitely go through the site and will let you know in case I get any issues.

Just to elaborate more, I am new to NLP area and pursuing research in NLP using Deep Learning so I need to test few existing Deep Learning based NLP research paper implementations on Hindi/Marathi language especially for Machine Translation (MT). For this, I need parallel sentence corpora for both the languages.