Workshop on Machine Learning in Speech and Language Processing

September 13, 2016
San Francisco, CA, USA


09:30-10:15Keynote talk: Kai YuStructured deep learning for context awareness in speech and language processing (slides)
10:15-11:00Keynote talk: Ian GoodfellowGenerative adversarial networks (slides)
11:20-11:40Spotlight invited talk: Ryan LoweModern challenges in learning end-to-end dialogue systems (slides)
11:40-12:00Spotlight invited talk: William ChanListen, attend and spell: A neural network for large vocabulary conversational speech recognition (slides)
12:00-12:20Spotlight invited talk: Herman KamperUnsupervised speech recognition using acoustic word embeddings (slides)
12:20-12:40Spotlight invited talk: Lisa Anne HendricksDeep compositional captioning: Describing novel object categories without paired training data (slides)
12:40-14:00Lunch break
14:00-14:45Keynote talk: Fei ShaBeing shallow and random is (almost) as good as being deep and thoughtful (slides)
14:45-15:30Keynote talk: Jan ChorowskiEnd-to-end approaches to speech recognition and language processing (slides)
15:30-17:00Break + poster session
Yossi Adi, Einat Kermany, Yonatan Belinkov, Ofer Lavi and Yoav GoldbergFine-Grained Analysis of Sentence Embeddings Using Auxiliary Prediction Tasks
Herman Kamper, Sharon Goldwater, and Aren JansenA Segmental Framework for Fully-Unsupervised Large-Vocabulary Speech Recognition
Ryan Lowe, Iulian Serban, Mike Noseworthy, Chia-Wei Liu, Laurent Charlin, and Joelle PineauAutomatically Evaluating Dialogue Responses
Muhammad Rizwan and David V. AndersonComparison of Distance Metrics for Phoneme Classification based on Deep Neural Network Features and Weighted k-NN Classifier
Parakrant Sarkar, Krothapalli Sreenivasa Rao, and Gurunath Reddy MDevelopment of Story Text-to-Speech System based on Story Genres
Sunil Thulasidasan and Jeffrey BilmesSemi-Supervised Phone Classification using Deep Neural Networks and Stochastic Graph-Based Entropic Regularization
Shubham Toshniwal and Karen LivescuRead, Attend and Pronounce: An Attention-Based Approach for Grapheme-To-Phoneme Conversion