TWiki> CSTR Web>Listen>ListenSemester1201617 (05 Dec 2016, Main.s1569548)EditAttach

2016-17Semester1
11.10.16Planning meeting
18.10.16Active and semi-supervised learning & optimising evaluation (Peter)
25.10.16

WaveNet: A Generative Model for Raw Audio (Blog post with examples). Background reading: Pixel Recurrent Neural Networks, Conditional Image Generation with PixelCNN Decoders, and Video Pixel Networks. (Steve)

01.11.16
Domain adversarial MTL - Practical paper from IS: Adversarial Multi-task Learning of Deep Neural Networks for Robust Speech Recognition and a background paper (JMLR): Domain-Adversarial Training of Neural Networks (Joachim)
08.11.16

Open-Domain Audio-Visual Speech Recognition: A Deep Learning Approach & Multimodal deep learning (Ondrej)

15.11.16
A DNN-HMM Approach to Story Segmentation and Dialogue Session Segmentation by Embedding-Enhanced TextTiling (Emiru)
22.11.16
Grouping context-dependent targets & phonetic context embeddings (Peter)
29.11.16
A review of Interspeech CNN/LSTM papers. Advances in Very Deep Convolutional Neural Networks for LVCSR & The IBM 2016 English Conversational Telephone Speech Recognition System Link to presentation (Joanna)
06.12.16
How neural network depth compensates for the conditional independence assumption (Joachim)
13.12.16
No meeting (SLT)
TBC
Speaker separation with deep clustering (Steve)
TBC
A mathematical theory of deep CNNs for feature extraction (Gustav)
TBC
Phone synchronous decoding with CTC lattice (Ondrej)
TBC
Generative adversarial networks review (Joanna and Joachim)
TBC
Review of Interspeech low-resource/semi-supervised papers
Topic revision: r9 - 05 Dec 2016 - 14:46:33 - Main.s1569548
 
This site is powered by the TWiki collaboration platformCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback
This Wiki uses Cookies