i used one wave file as input to the wavfile demo, lattice demo and transcriber demo. but i got the wrong output only.. the output is as follows
F:\sphinx4\sphinx4>java -jar bin/WavFile.jar
Loading Recognizer as defined in 'jar:file:/F:/sphinx4/sphinx4/bin/WavFile.jar!/
edu/cmu/sphinx/demo/wavfile/config.xml'...
Decoding jar:file:/F:/sphinx4/sphinx4/bin/WavFile.jar!/edu/cmu/sphinx/demo/wavfi
le/ls1b.wav
Result: oh two eight seven oh oh oh two oh eight oh five six two five eight oh o
h eight six eight eight four zero six eight six oh eight nine six eight oh one o
h eight eight oh eight oh six eight oh one oh two nine six oh one eight eight fi
ve eight two eight oh six eight eight eight three eight five two eight eight two
three eight eight six oh oh two five eight two oh eight six oh two six oh eight
nine six eight oh one eight oh six oh one oh two nine six two eight oh six eigh
t eight five oh two two one oh eight one two two eight oh eight two eight oh fou
r eight eight four eight oh two eight two eight oh three six two five two five o
h two eight oh oh oh two oh eight oh five oh two five oh oh oh five five five si
x two five eight oh five six five eight oh eight six eight eight oh eight oh two
oh two eight oh five six eight oh five eight two two eight two eight two two ei
ght two eight eight two eight four zero seven six oh three eight oh four one eig
ht oh oh eight oh four two two eight oh oh two one two two one eight oh three ei
ght eight six oh oh two five oh two oh five five six two five six oh eight six o
h eight oh two five six eight oh two eight two two eight two eight oh six four z
ero seven six eight oh four eight oh four two two one two two eight oh two one e
ight oh eight eight oh eight oh four eight oh eight oh two nine three two five f
our eight oh eight eight oh eight eight eight eight one
========================================
please help me. how to get the correct output..
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
You do know that the grammar is setup to only recognize 0-9 right?
If you wanted it to recognize other stuff, change the JSGF grammar below and add more to the acoustic model/dict.
/*
* JSGF Digits Grammar for Hello World example /
grammar digits;
public <numbers> = (oh | zero | one | two | three | four | five | six | seven | eight | nine) * ;
Go into more detail of what the "correct output" should be.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
the ls1b.wav contains the spoken english class details. i want to retrieve the words. so please tell me how to change the grammer setup in the config.xml file .
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
hi..
i used one wave file as input to the wavfile demo, lattice demo and transcriber demo. but i got the wrong output only.. the output is as follows
F:\sphinx4\sphinx4>java -jar bin/WavFile.jar
Loading Recognizer as defined in 'jar:file:/F:/sphinx4/sphinx4/bin/WavFile.jar!/
edu/cmu/sphinx/demo/wavfile/config.xml'...
Decoding jar:file:/F:/sphinx4/sphinx4/bin/WavFile.jar!/edu/cmu/sphinx/demo/wavfi
le/ls1b.wav
Result: oh two eight seven oh oh oh two oh eight oh five six two five eight oh o
h eight six eight eight four zero six eight six oh eight nine six eight oh one o
h eight eight oh eight oh six eight oh one oh two nine six oh one eight eight fi
ve eight two eight oh six eight eight eight three eight five two eight eight two
three eight eight six oh oh two five eight two oh eight six oh two six oh eight
nine six eight oh one eight oh six oh one oh two nine six two eight oh six eigh
t eight five oh two two one oh eight one two two eight oh eight two eight oh fou
r eight eight four eight oh two eight two eight oh three six two five two five o
h two eight oh oh oh two oh eight oh five oh two five oh oh oh five five five si
x two five eight oh five six five eight oh eight six eight eight oh eight oh two
oh two eight oh five six eight oh five eight two two eight two eight two two ei
ght two eight eight two eight four zero seven six oh three eight oh four one eig
ht oh oh eight oh four two two eight oh oh two one two two one eight oh three ei
ght eight six oh oh two five oh two oh five five six two five six oh eight six o
h eight oh two five six eight oh two eight two two eight two eight oh six four z
ero seven six eight oh four eight oh four two two one two two eight oh two one e
ight oh eight eight oh eight oh four eight oh eight oh two nine three two five f
our eight oh eight eight oh eight eight eight eight one
========================================
please help me. how to get the correct output..
Read the documentation and tutorials on CMU's site. You will have difficulty building such a system without reading it.
What is inside that wav ls1b.wav?
You do know that the grammar is setup to only recognize 0-9 right?
If you wanted it to recognize other stuff, change the JSGF grammar below and add more to the acoustic model/dict.
/*
* JSGF Digits Grammar for Hello World example
/
grammar digits;
public <numbers> = (oh | zero | one | two | three | four | five | six | seven | eight | nine) * ;
Go into more detail of what the "correct output" should be.
thank u for reply sir,
the ls1b.wav contains the spoken english class details. i want to retrieve the words. so please tell me how to change the grammer setup in the config.xml file .