Start with HTML::Parser (or one of its relatives like HTML::Filter). Writing a decent HTML parser inside a regular-expression is impossible.

package HTML::Quoter;
require HTML::Filter;
@ISA=qw(HTML::Filter);

my $data='';

sub output{
  my $self=shift;
  my $d=$_[0];
  if($d=~/\<\s*\/?\s*(\w+)/){
    $data.=$d;
  }else{
    $d=~s/\"/&quot;/gs;
    $data.=$d;
  }
}

my $p=HTML::Quoter->new();
$p->parse_file("quotes.html");
print $data;
[download]