Let's say I have a simple 3-gram LM with only 10 words: zero, one, two, three,
four, five, six, seven, eight, nine.
And a phonetic dictionary with 12 entries:
zero <pronuntiation1>
one(1) <pronuntiation1>
one(2) <pronuntiation2>
one(3) <pronuntiation3>
two <pronuntiation1>
...
nine <pronuntiation1> </pronuntiation1></pronuntiation1></pronuntiation3></pronuntiation2></pronuntiation1></pronuntiation1>
Will the decoder use all the phonetic pronuntiations for word "one" or the
round brackets are treated as usual characters and the decoder will say: "I
don't have the word <one(1)> in the LM, therefore I won't include it in the
search graph"? </one(1)>
Horia
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi,
Let's say I have a simple 3-gram LM with only 10 words: zero, one, two, three,
four, five, six, seven, eight, nine.
And a phonetic dictionary with 12 entries:
zero <pronuntiation1>
one(1) <pronuntiation1>
one(2) <pronuntiation2>
one(3) <pronuntiation3>
two <pronuntiation1>
...
nine <pronuntiation1> </pronuntiation1></pronuntiation1></pronuntiation3></pronuntiation2></pronuntiation1></pronuntiation1>
Will the decoder use all the phonetic pronuntiations for word "one" or the
round brackets are treated as usual characters and the decoder will say: "I
don't have the word <one(1)> in the LM, therefore I won't include it in the
search graph"? </one(1)>
Horia
This is the case where it's easier to try than to ask ;)