pruned_transducer_stateless7_streaming model finetuning problem

[younghounSon]

I did fine tuning like the image above, but it is not recognized continuously. It is recognized only once and after that, the microphone input is ignored.
Before fine tuning, it worked fine even after recognizing the utterance once.