Fixes to #3295 Improve RNN-T streaming decoding (#3379)

Summary:
Pull Request resolved: https://github.com/pytorch/audio/pull/3379

Fixes `RNNTBeamSearch.infer`'s docstring and removes unused import from tutorial.

Reviewed By: mthrok

Differential Revision: D46227174

fbshipit-source-id: 7c1c3f05a6476cb0437622dea6f3ae6cb3ea9468

examples/tutorials/online_asr_tutorial.py

@@ -39,7 +39,6 @@ to perform online speech recognition.
 # --------------
 #
 
-import os
 import torch
 import torchaudio
 

torchaudio/models/rnnt_decoder.py

@@ -314,7 +314,7 @@ class RNNTBeamSearch(torch.nn.Module):
             state (List[List[torch.Tensor]] or None, optional): list of lists of tensors
                 representing transcription network internal state generated in preceding
                 invocation. (Default: ``None``)
-            hypothesis (Hypothesis or None): hypothesis from preceding invocation to seed
+            hypothesis (List[Hypothesis] or None): hypotheses from preceding invocation to seed
                 search with. (Default: ``None``)
 
         Returns: