Geonmin Kim

이전에 분리했던 텍스트들에 따라 문장분리 결과가 달라지는 사례

4

* split_sentences()호출시 이전에 처리했던 텍스트에 따라서 현재 처리하는 텍스트가 달라질 요소가 있나요? * 있다면 어떤 옵션으로 방지할 수 있을까요? * kss version: 3.3.1.1 * split_sentences(text, backend="mecab", num_workers=1) * text =...

Oracle entity in Table 2 VS. Oracle keywords in Table 7

5

I am trying to reproduce ROUGE on CNNDM with **'oracle keyword in Table 7'**. **'oracle entity setting in Table 2'** sounds similar to **'oracle keyword in Table 7'**, however, ROUGE...

Why gradient does not scale linearly with CTCloss in pytorch binding?

2

I am pytorch user, and I am curious this issue is generally occurs in every platform of CTC. Originally, this issue is posted in https://github.com/SeanNaren/deepspeech.pytorch/issues/250. I am trying to use...

Thanks for sharing interesting, practical work. I am thinking about replacing BN to proposed WN layer for speech GAN. Is it possible to apply 'WeightNormalizedConv2d' & 'WeightNormalizedConvTranspose2d' layer to 1d...

First of all, thanks for sharing source code for your awesome work. I am trying to apply REINFORCE to train my model, and trying to use idea of reducing variance...

p(blank symbol) >> p(non-blank symbol) during NN-CTC training

1

Hi all I want to discuss some issue regarding training DNN/CNN-CTC for speech recognition. (Wall Street Journal Corpus). I modeled output unit as characters. I observed that CTC objective function...

How to test pbrakels's version CTC?

4

Hi all ! I am planning to merge CTC with KALDI for building end-to-end speech recognition system. First of all, thanks for providing CTC as open source. I attribute yours...

Questions regarding detecting sentence boundary from paragraph

2

Hi ! Thanks to sharing your work. I wonder how you divide sentences from paragraphs of 'HotelReview Corpus'. I guess you divide sentence boundary based on punctuation marks such as...

Questions on Loss scale, Hyperparameters

4

Thanks to share easy-to-follow code. I am currently applying WGAN to learning text distribution. Here is questions regarding WGAN. Question1. In Figure 3, the loss of MLP and DCGAN seems...

Hi ! Thanks for sharing your awesome source code. I enjoy performance of this implementation. I wonder your opinion about batch normalization. There are various issues applying batch normalization for...