Registration site is closed, but you can register on site!

Welcome to the web site of The First International Workshop on Machine Learning in Spoken Language Processing (MLSLP2015) held at The University of Aizu, Aizu, Fukushima, Japan, on September 19-20, 2015, which is co-hosted by ASJ & IEICE SIG-SP, and co-sponsored by The University of Aizu, IEEE SPS Japan Chapter, IEEE SPS Kansai Chapter, and IEEE SPS Sendai Chapter.

Recent dramatic advances of spoken language technologies have made use of highly sophisticated machine learning techniques such as deep learning, RNN, LSTM and many other neural networks approaches. As spoken language technologies span very broad areas such as acoustic modeling, language modeling, dialogue management, and speech synthesis, we hope that this workshop can gather researchers to share their findings and expertise in these related fields.

We welcome contributions in the following topics (but not limited to):

  • Acoustic modeling based on DNN
  • Language modeling based on RNN
  • Discriminative training
  • Adaptation/normalization
  • Speech synthesis based on statistical models
  • Model-based speech conversion
  • Speaker recognition using i-vector/UBM
  • Statistical dialogue management
  • Topic model
  • Statistical speech translation
  • Speech mining

The workshop will closely cooperate with the IEICE Transactions on Information and Systems and a special section of the same theme is planned for October 2016 issue (submission deadline is January 27, 2016). Authors of accepted papers will be strongly recommended to submit extended papers to the transactions.