Intercept statement recognition during parsing

[Korporal]

Is there a way to intercept the detection of a valid rule, in such a way that we can then adjust a property used by the lexer? Consider:

@lexer::members {
    
	// This code MUST compile fine as Java and C#

	public String KeywordLexiconCode = "en";

	public Boolean Lexicon(String Code)
	{
		return Code == KeywordLexiconCode;
	}
}

Is there a way - during the parse - for me to have the parser recognize some rule and at that point modify the string KeywordLexiconCode thus influencing the remainder of the tokenization and parsing? The string is used in token predicates like this:

INTERRUPT: 
           {Lexicon("bg")}? ('interrupt') | 
           {Lexicon("de")}? ('unterbrechen') | 
           {Lexicon("en")}? ('interrupt') | 
           {Lexicon("fr")}? ('interrompre') | 
           {Lexicon("he")}? ('interrupt') | 
           {Lexicon("nl")}? ('interrupt') | 
           {Lexicon("ru")}? ('interrupt') ;

I know we have parse tree visitors and listeners, but not sure if these are an appropriate place to do this.