This object implements an n-gram language model.
Public Methods
-
void readARPA( FILE* arpa_fd )
- Creates a language model from an ARPA format file.
Public Members
-
Creates the language model.
-
Calculates the language model probability (with backoff) of the word
-
Calculates the language model probability (with backoff) of 'next_word'
-
given the previous word sequence stored in the hypothesis pointed to
-
Creates a language model from an Noway binary LM format file.
Documentation
This object implements an n-gram language model. The n-gram data structures
are encapsulated in the ngram member variable (see LMNGram class). Methods
are provided to read a LM file in ARPA format or in Noway binary format.
A method is provided to calculate a LM prob (with backoff) for a given
sequence of words.
- Creates the language model.
- Creates the language model.
- Calculates the language model probability (with backoff) of the word
- Calculates the language model probability (with backoff) of the word
- Calculates the language model probability (with backoff) of 'next_word'
- Calculates the language model probability (with backoff) of 'next_word'
- given the previous word sequence stored in the hypothesis pointed to
- given the previous word sequence stored in the hypothesis pointed to
- void readARPA( FILE* arpa_fd )
- Creates a language model from an ARPA format file. Internal function.
- Creates a language model from an Noway binary LM format file.
- Creates a language model from an Noway binary LM format file. Internal function.
- This class has no child classes.
- Author:
- Darren Moore (moore@idiap.ch)
Alphabetic index HTML hierarchy of classes or Java
This page was generated with the help of DOC++.