[Corpora-List] n-best/lattice hypothesis for speech recognition

[Dharmendra Kanejiya]

Dear list members,

I am developing a statistical langauge model using long distance LSA
semantics and syntax. I have tested it for perplexity evaluation and now
would like to test for speech recognition word error rate. But we don't have
any speech recognition engine for this, so i would like to test it as a
second pass language model for rescoring hypthesis. For this, i would
greatly appreciate if someone can point me to the appropriate sources from
where i can obtain the first pass n-best/lattice hypthesis.

Thanks in advance,

Dharmendra Kanejiya
PhD student
Indian Institute of Technology Delhi
http://www.cse.iitd.ernet.in/~eer99010


_________________________________________________________________
MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*.
http://join.msn.com/?page=features/virus