Adding an example of using pretrained model to predict emotion in local audio file

Update README.md

Update README.md

Update README.md

Update README.md

Update README.md

Create README.md