Support for Text-to-Speech models

Optimum Intel is a great tool that combines the convenience of HF API with the performance of OpenVINO. It helps build virtual assistants with one exception - Text-to-Speech pipelines. Hence, I'm requesting that support be added, which will help include speech generation in our AI Ref Kits promoting Optimum Intel and OpenVINO.

Example
```
from optimum.intel import pipeline

synthesizer = pipeline("text-to-speech", "suno/bark")

synthesizer("Look I am generating speech in three lines of code!")
```