- Spanish: Simone, Eliza, Bridget, Jimmy, Naomi
- German: Alyssa, Jacob, Andrea, Shibo
- Korean: Chloe, Hankyul, Shiven, Shohini
If anyone is missing from this list, please join one of the groups!
If anyone is missing from this list, please join one of the groups!
Based on the documentation:
This operation randomly generates a set of successful paths in the input FST. The operation relies on an
ArcSelector
object for randomly selecting an outgoing transition at a given state in the input FST. The default arc selector,UniformArcSelector
, randomly selects a transition using the uniform distribution.LogProbArcSelector
randomly selects a transition w.r.t. the weights treated as negative log probabilities after normalizing for the total weight leaving the state. In all cases, finality is treated as a transition to a super-final state.
_____________________________________
Example:
fstrandgen G.fst rand1.fst
fstprint --acceptor --isymbols=words.txt rand1.fst
0 1 WILL
1 2 WABASH
2 3 SEVENTEENTH
3 4 OCTOBER
4 5
5
fstdraw --acceptor --isymbols=words.txt rand1.fst | dot -Tx11
My demo is under /projects/speech/sys/kaldi-master/egs/rm/s5-sb2295/demo/fstrandgen
and you can run it with source demo.sh
The Bengali speech corpus that we have on kay is Shruti. It includes the transcriptions and the phonetic dictionary. More information about it can be found here: http://cse.iitkgp.ac.in/~pabitra/shruti_corpus.html
UPDATE: We have another Bengali corpus now: IARPA Babel Bengali