Segment models are a generalization of HMMs that can represent feature dynamics and/or correlation in time. In this work we develop the theory of Bayesian and maximum-likelihood adaptation for a segment model characterized by a polynomial mean trajectory. We show how adaptation parameters can be shared and adaptation detail can be controlled at run-time based on the amount of adaptation data av...

We propose a novel generative neural network architecture for Dialogue Act classification. Building upon the Recurrent Neural Network framework, our model incorporates a new attentional technique and a label-to-label connection for sequence learning, akin to Hidden Markov Models. Our experiments show that both of these innovations enable our model to outperform strong baselines for dialogue-act...

This paper describes a noisy channel model of speech repairs, which can identify and correct repairs in speech transcripts. A syntactic parser is used as the source model, and a novel type of TAG-based transducer is the channel model. The use of TAG is motivated by the intuition that the reparandum is a “rough copy” of the repair. The model is trained and tested on the Switchboard disfluency-an...

This paper describes a time-series model for parsing transcribed speech containing disfluencies. This model differs from previous parsers in its explicit modeling of a buffer of recent words, which allows it to recognize repairs more easily due to the frequent overlap in words between errors and their repairs. The parser implementing this model is evaluated on the standard Switchboard transcrib...

This paper describes the 2000 BBN Byblos Large Vocabulary Continuous Speech Recognition (LVCSR) system. We briefly outline the training and decoding procedures used in the system, and explain in detail the new features we have added to the system in the past year. These new features include multiple adaptation stages, parallel path rescoring, and a new word confidence system. Word error rate re...

The past decade has seen tremendous progress in experimentally realizing the building blocks of quantum repeaters. Repeater architectures with multiplexed memories have been proposed to increase entanglement distribution rates, but an open challenge is maintain fidelity over long-distance links. Here, we address this a router architecture comprising many connected photonic switchboard broker fl...

Abstract Dialog acts can be interpreted as the atomic units of a conversation, more fine-grained than utterances, characterized by specific communicative function. The ability to structure conversational transcript sequence dialog acts—dialog act recognition, including segmentation—is critical for understanding dialog. We apply two pre-trained transformer models, XLNet and Longformer, this task...

This paper applies the recently proposed SPAM models for acoustic modeling in a Speaker Adaptive Training (SAT) context on large vocabulary conversational speech databases, including the Switchboard database. SPAM models are Gaussian mixture models in which a subspace constraint is placed on the precision and mean matrices (although this paper focuses on the case of unconstrained means). They i...

We describe a new objective for graph-based semi-supervised learning based on minimizing the Kullback-Leibler divergence between discrete probability measures that encode class membership probabilities. We show how the proposed objective can be efficiently optimized using alternating minimization. We prove that the alternating minimization procedure converges to the correct optimum and derive a...