N-best error simulation for training spoken dialogue systems

A recent trend in spoken dialogue research is the use of reinforcement learning to train dialogue systems in a simulated environment. Past researchers have shown that the types of errors that are simulated can have a significant effect on simulated dialogue performance. Since modern systems typically receive an N-best list of possible user utterances, it is important to be able to simulate a full N-best list of hypotheses. This paper presents a new method for simulating such errors based on logistic regression, as well as a new method for simulating the structure of N-best lists of semantics and their probabilities, based on the Dirichlet distribution. Off-line evaluations show that the new Dirichlet model results in a much closer match to the receiver operating characteristics (ROC) of the live data. Experiments also show that the logistic model gives confusions that are closer to the type of confusions observed in live situations. The hope is that these new error models will be able to improve the resulting performance of trained dialogue systems. © 2012 IEEE.

Complementary Phone Error training

This paper introduces a novel method for the training of a complementary acoustic model with respect to set of given acoustic models. The method is based upon an extension of the Minimum Phone Error (MPE) criterion and aims at producing a model that makes complementary phone errors to those already trained. The technique is therefore called Complementary Phone Error (CPE) training. The method is evaluated using an Arabic large vocabulary continuous speech recognition task. Reductions in word error rate (WER) after combination with a CPE-trained system were obtained with up to 0.7% absolute for a system trained on 172 hours of acoustic data and up to 0.2% absolute for the final system trained on nearly 2000 hours of Arabic data.

Vibration from railways: Can we achieve better than +/-10dB prediction accuracy?

This paper examines the sources of uncertainly in models used to predict vibration from underground railways. It will become clear from this presentation that by varying parameters by a small amount, consistent with uncertainties in measured data, the predicted vibration levels vary significantly, often by more than 10dB. This error cannot be forecast. Small changes made to soil parameters (Compressive and Shear Wave velocities and density), to slab bending stiffness and mass and to the measurement position give rise to changes in vibration levels of more than lOdB. So if 10dB prediction error results from small uncertainties in soil parameters and measurement position it cannot be sensible to rely on prediction models for accuracy better than 10dB. The presentation will demonstrate in real time the use of the new - and freely-available - PiP software for calculating vibration from railway tunnels in real time.

A derivation of the source-channel error exponent using nonidentical product distributions

This paper studies the random-coding exponent of joint source-channel coding for a scheme where source messages are assigned to disjoint subsets (referred to as classes), and codewords are independently generated according to a distribution that depends on the class index of the source message. For discrete memoryless systems, two optimally chosen classes and product distributions are found to be sufficient to attain the sphere-packing exponent in those cases where it is tight. © 2014 IEEE.