Masao Someki

@1nlplearner At the first execution, onnxruntime takes a longer time for inference. So please skip the first execution.

Thank you @popcornell, I think it is fine for just using the lhotze as a dataset. And as for the augmentations, you are right, we cannot take its advantage in...

@popcornell > I think current ESPNet for example at least feeds the data that has been speed augmented no ? You are right, I overlooked this point!! Then I think...

With the partially auto-regressive decoding we can speed up OWSM inference twice as fast as Whisper on Google colab. We can use PAR inference with the following code: ```python from...

The paper encountered a memory issue, but a minor change in attention ([commit](ee641aa63714ab0135846da2c8fae77d7e87eb5f)) resolved this problem. Additionally, it contributed to speeding up the inference process.

Thank you @pyf98! > BTW, I didn't read the paper. Why does it only support ASR? Does it rely on CTC? If not, ASR and ST would be similar? Also,...

@pyf98 I noticed that there were quite a lot of noisy codes I used in my experiments, so I removed everything unrequired. I also added comments and re-formated codes.

Hi, when I try to run the integration test for the `gan-tts` task, I encounter the following error. I'm using the `egs2/mini_an4/tts/conf/train_vits_debug.yaml` to test the training. Does anyone know how...

I also encountered the following error when I tried to run the ENF-TSE task. ```plain Traceback (most recent call last): File "../../../test/espnetez/test_integration_espnetez.py", line 361, in trainer.train() File "/hdd/doc/espnet/espnetez/trainer.py", line 170,...

@pyf98 @pengchengguo Thank you both for testing this PR!! I now understand what is happening with this issue. The inefficient memory was, in fact, the result of the cache created...