Order spokenforms Sections

For characters, abbreviations, file name extenstions and punctuation there are also provisions in the spokenforms module. They are stored in the next set of instance variables.

 
    char2spoken = {} # characters of radio alphabet possibly extended, got from [alphabet] section
    spoken2char = {}
    abbrev2spoken = {} # lists of spoken forms, got from [abbrevs] sections
    spoken2abbrev = {}
    ext2spoken = {}    # file extensions (without the dot) (got from [extensions] section)
    spoken2ext = {}
    punct2spoken = {}  # punctuation, from section [punctuationreverse] (other way round!)
    spoken2punct = {}

Characters

As can be seen in the grammar _keystrokes, the radio alphabet is used via the [alphabet] section of the xxx_spokenforms.ini file.

The grammar list {character} is filled with a call to the function self.setCharactersList('character').

At recognition time, the value (character) is retrieved with the function self.getCharacterFromSpoken(). See _keystrokes.py.

Punctuation

Also in the grammar _keystrokes the punctuation is used. The functions are self.setPunctuationList('punctuation') and self.getPunctuationFromSpoken().

These four functions are defined in the IniGrammar class in natlinkutilsbj.py.