Hyejeong Jo

Have you ever solved this problem?

@upskyy thank you. I fix that problem. And another error was revealed, File ~/bin/miniconda3/envs/EEGToText/lib/python3.8/site-packages/openspeech/__init__.py:23 1 # MIT License 2 # 3 # Copyright (c) 2021 Soohwan Kim and Sangchun Ha...

Oh I found. I install libsora==0.9.1, that problem is solved. thank you

@csukuangfj Thank you. I said that in a misleading way. What I'm curious about is why target_length +1 needs to be entered as the RNNT loss's 3rd input. Looking at...

So the number of classes should be len(vocab)? I understand. I had misunderstood the mechanism of RNN-Transducer. Since model will start from a blank label, it should be target_length+1.

@csukuangfj Thank you for your kindness.

I found how to fix this problem. In eval_decoding.py file, `predictions=tokenizer.encode(predicted_string)` this code should be changed to `predictions=tokenizer.encode(predicted_string[0])` predicted_string is list, so we put the only string. In our case,...