Hybrid speech recognition with Deep Bidirectional LSTM

Abstract

Data analytics helps basketball teams to create tactics. However, manual data collection and analytics are costly and ineffective. Therefore, we applied a deep bidirectional long short-term memory (BLSTM) and mixture density network (MDN) approach. This model is not only capable of predicting a basketball trajectory based on real data, but it also can generate new trajectory samples. It is an excellent application to help coaches and players decide when and where to shoot. Its structure is particularly suitable for dealing with time series problems. BLSTM receives forward and backward information at the same time, while stacking multiple BLSTMs further increases the learning ability of the model. Combined with BLSTMs, MDN is used to generate a multi-modal distribution of outputs. Thus, the proposed model can, in principle, represent arbitrary conditional probability distributions of output variables. We tested our model with two experiments on three-pointer datasets from NBA SportVu data. In the hit-or-miss classification experiment, the proposed model outperformed other models in terms of the convergence speed and accuracy. In the trajectory generation experiment, eight model-generated trajectories at a given time closely matched real trajectories.

Document Modeling with Gated Recurrent Neural Network for Sentiment Classification

“…To address this problem, we develop a gated recurrent neural network for document composition, which works in a sequential way and adaptively encodes sentence semantics in document representations. The approach is analogous to the recently emerged LSTM (Graves et al, 2013;Zaremba and Sutskever, 2014;Xu et al, 2015) and gated neural network (Cho et al, 2014;Chung et al, 2015). Specifically, the transition function of the gated RNN used in this work is calculated as follows.…”

Neural Modeling of Multi-Predicate Interactions for Japanese Predicate Argument Structure Analysis

“…The single-sequence model exploits stacked bidirectional RNNs (Bi-RNN) (Schuster and Paliwal, 1997;Graves et al, 2005Graves et al, , 2013Zhou and Xu, 2015). Figure 3 shows the overall architecture, which consists of the following three components:…”

When and why are log-linear models self-normalizing?

“…On a variety of tasks, these LMs have produced substantial gains over conventional generative models based on counting n-grams. Successes include machine translation (Devlin et al, 2014) and speech recognition (Graves et al, 2013). However, log-linear LMs come at a significant cost for computational efficiency.…”