change speechbrain.pretrained to speechbrain.inference.separation to avoid import error

when running the current code, we get the following error:
```
from speechbrain.pretrained import SepformerSeparation as separator
2 import torchaudio
4 model = separator.from_hparams(source="speechbrain/sepformer-wham16k-enhancement", savedir='pretrained_models/sepformer-wham16k-enhancement')

ModuleNotFoundError: No module named 'speechbrain.pretrained'
```
by referring to this [issue](https://github.com/speechbrain/speechbrain/issues/2472), we can find that change speechbrain.pretrained to speechbrain.inference.separation solve that error

README.md

@@ -43,7 +43,7 @@ Please notice that we encourage you to read our tutorials and learn more about [
 ### Perform speech enhancement on your own audio file
 
 ```python
-from speechbrain.pretrained import SepformerSeparation as separator
+from speechbrain.inference.separation import SepformerSeparation as separator
 import torchaudio
 
 model = separator.from_hparams(source="speechbrain/sepformer-wham16k-enhancement", savedir='pretrained_models/sepformer-wham16k-enhancement')