This should be straightforward - I wouldn't use the id of the record type, as that can change between sandbox and production when you deploy it.  Instead pull the record type back based on its name.  In the code below, the name is assumed to be 'My Account RT' - change this with the name of your account record type:

before line 7:

// get the record type 
RecordType rt=[select id from RecordType where Name='My Account RT' and sobjectype='Account'];

// build a set (no dupes) of the account ids associated with the contacts
Set<id> accIds=new Set<accIds>();
for (Contact cont : trigger.new)
{
  accIds.add(cont.accountId);
}

// pull back all accounts with the record type that allows dupes
List<Account> accs=[select id from Account where RecordTypeId=:rt.id and id in :accIds];

// build a map of account that allows dupes by id
Map<Id, Account> accsById=new Map<Id, Account>();
accsById.putAll(accs);

Then wrap lines 9 - 17 in the following if statement - this will ignore any contacts that are created for accounts with the special record type.

if (null==(accsById.get(contact.AccountId))
{
}

Finally in line 23, change the query to exclude contacts from the special record type when retrieving duplicates:

for (Contact contact : [select Email from Contact where Email in :contactMap.keySet() and AccountId not in :accsById.keySet()]