ECE 590SIP, Fall 2019, will be a series of lectures on speech technology.



  1. Aug 26: A short history of speech technology, Mark Hasegawa-Johnson
  2. Sep 9: Speaker Adaptation of Neural Networks with Learning Speaker Aware Offsets, Leda Sari
  3. Sep 16, 10:00-11:00 AM AutoGAN: Neural Architecture Search for Generative Adversarial Networks, Zhangyang (Atlas) Wang
  4. Sep 23: A Comparable Phone Set for the TIMIT Dataset Discovered in Clustering of Listen, Attend and Spell, Jialu Li
  5. Sep 30: AutoVC: A Zero-Shot Style Transfer with Only Autoencoder Loss, Kaizhi Qian
  6. Oct 7: Unsupervised Deep Clustering for Source Separation, Shrikant Venkataramani. Slides, article.
  7. Oct. 14: How to build your own supercomputer for training and testing deep neural networks. Partly following Jeff Chen's tutorial, partly using this list of parts picked from PC Part Picker, which is important to guarantee that the parts are all compatible.
  8. Oct. 21: Capturing L1 Influence on L2 Pronunciation by Simulating Perceptual Space Using Acoustic Features, Shuju Shi.