Describe the bug

This issue serves as a placeholder to hold the discussion and progress of tracking the compatibility of SpeechBrain with ONNX.

We have many issues / discussions of people trying to export our models with ONNX, yet we don't have a proper resource to answer this. We need to clarify this somewhere (tutorial? documentation?) and maybe provide a map of the main models we have with their compatibility to ONNX ? Typically ECAPA, ASR (SSL, Conformer), VAD, TTS, Translation?

Expected behaviour

Add documentation to support ONNX exports.

To Reproduce

No response

Environment Details

No response

Relevant Log Output

No response

Additional Context

No response