# Tokenise
my @tokens = $doc =~ /<[^>]*>|./sg;

# Extract text
my $text = '';
for (@tokens) {
   if    (/^<([0-9]+)>/) { $text .= chr($1); }
   elsif (/^[^<]/)       { $text .= $1;      }
}

# Do stuff with $text
# ...

# Print resulting text
print join('', $text);