In this paper, we apply a bidirectional Long Short-Term Memory with a Conditional Random Field to the task of disfluency detection. Long-range dependencies is one of the core problems for disfluency detection. Our model handles long-range dependencies by both using the Long Short-Term Memory and hand-crafted discrete features. Experiments show that utilizing the hand-crafted discrete features s...

We are concerned with robust and accurate forecasting of multiphase flow rates in wells and pipelines during oil and gas production. In practice, the possibility to physically measure the rates is often limited; besides, it is desirable to estimate future values of multiphase rates based on the previous behavior of the system. In this work, we demonstrate that a Long Short-Term Memory (LSTM) re...

For real-world driver drowsiness detection from videos, the variation of head pose is so large that the existing methods on global face is not capable of extracting effective features, such as looking aside and lowering head. Temporal dependencies with variable length are also rarely considered by the previous approaches, e.g., yawning and speaking. In this paper, we propose a Longterm Multi-gr...

In this work, Portuguese, Polish, English, Urdu, and Arabic automatic speech recognition evaluation systems developed by the RWTH Aachen University are presented. Our LVCSR systems focus on various domains like broadcast news, spontaneous speech, and podcasts. All these systems but Urdu are used for Euronews and Skynews evaluations as part of the EUBridge project. Our previously developed LVCSR...

Curriculum Learning emphasizes the order of training instances in a computational learning setup. The core hypothesis is that simpler instances should be learned early as building blocks to learn more complex ones. Despite its usefulness, it is still unknown how exactly the internal representation of models are affected by curriculum learning. In this paper, we study the effect of curriculum le...

Transmembrane Protein Prediction is a problem with many uses as experimental determination of protein structures is still expensive and for different purposes it can be useful to know the structure. Here I introduce a small long short-term memory network based model which gives a precision of 67 ± 3 and a recall of 71± 3. The model manages, when compared to TMSEG [3], slightly worse but is stil...

Computational models for sarcasm detection have often relied on the content of utterances in isolation. However, speaker’s sarcastic intent is not always obvious without additional context. Focusing on social media discussions, we investigate two issues: (1) does modeling of conversation context help in sarcasm detection and (2) can we understand what part of conversation context triggered the ...

Speaker identification refers to the task of localizing the face of a person who has the same identity as the ongoing voice in a video. This task not only requires collective perception over both visual and auditory signals, the robustness to handle severe quality degradations and unconstrained content variations are also indispensable. In this paper, we describe a novel multimodal Long Short-T...

We investigate fully automatic recognition of singer traits, i. e., gender, age, height and ‘race’ of the main performing artist(s) in recorded popular music. Monaural source separation techniques are combined to simultaneously enhance harmonic parts and extract the leading voice. For evaluation the UltraStar database of 581 pop music songs with 516 distinct singers is chosen. Extensive test ru...

In this paper, we present our system which is evaluated in the TREC 2016 LiveQA Challenge. Same as the last year, the TREC 2016 LiveQA track focuses on “live” question answering for the real-user questions from Yahoo! Answer. In this year, we first apply a parameter sharing Long Short Term Memory(LSTM) network to learn a high embedding of question representation. Then we combine the question re...