Text this: Improving LLM Outputs Against Jailbreak Attacks With Expert Model Integration