Extract hidden states of PyTorch module #606

Hi,
First of all, thank you for this brilliant framework.

Is it possible to extract model representations of the pre-trained models for a given input. Say I load:

from speechbrain.pretrained import EncoderDecoderASR
asr_model = EncoderDecoderASR.from_hparams(source="speechbrain/asr-crdnn-rnnlm-librispeech",
											 savedir="pretrained_models/asr-crdnn-rnnlm-librispeech")
asr_model.transcribe_file('speechbrain/asr-crdnn-rnnlm-librispeech/example.wav')
asr_model.modules

Under asr_model.modules I am able to see the PyTorch modules, but I am not able to see the hidden states (similar to Huggingface model(output_hidden_states=True). Is this currently possible in Speechbrain?

Thank you,
Greta