In this paper, we present a taxonomy of stories told in dialogue. We based our scheme on prior work analyzing narrative structure and method of telling, relation to storyteller identity, as well as some categories particular to dialogue, such as how the story gets introduced. Our taxonomy currently has 5 major dimensions, with most having sub-dimensions each dimension has an associated set of d...

In this paper we report our recent research whose goal is to improve the performance of a novel speech recognizer based on an underlying statistical hidden dynamic model of phonetic reduction in the production of conversational speech. We have developed a path-stack search algorithm which efficiently computes the likelihood of any observation utterance while optimizing the dynamic regimes in th...

We investigate how the probability of a word affects its pronunciation. We examined 5618 tokens of the 10 most frequent (function) words in Switchboard: I, and, the, that, a, you, to, of, it, and in, and 2042 tokens of content words whose lexical form ends in a t or d. Our observations were drawn from the phonetically hand-transcribed subset [1] of the Switchboard corpus [2], enabling us to cod...

Recently, an extension to standard hidden Markov models for speech recognition called Hidden Model Sequence (HMS) modelling was introduced. In this approach the relationship between phones used in a pronunciation dictionary and the HMMs used to model these in context is assumed to be stochastic. One important feature of the HMS framework is the ability to handle arbitrary model to phone sequenc...

In this paper, we describe the ISIP Automatic Speech Recognition system (ISIP-ASR) used for the Hub-5 2000 English evaluations. The system is a public domain cross-word context-dependent HMM based system and has all the functionality normally expected in an LVCSR system, including Baum-Welch training for continuous density HMMs, phonetic decision tree-based state-tying, word graph generation an...

This paper documents the use of conversational telephone speech test materials in the NIST coordinated evaluation conducted early in 2000. The primary evaluation was of General American English speech, but a subsidiary evaluation of Mandarin speech was also offered. The primary test data consisted of twenty conversations collected for the original Switchboard Corpus but not released with the pu...

We describe the 2017 version of Microsoft’s conversational speech recognition system, in which we update our 2016 system with recent developments in neural-network-based acoustic and language modeling to further advance the state of the art on the Switchboard speech recognition task. The system adds a CNN-BLSTM acoustic model to the set of model architectures we combined previously, and include...

Recent developments in statistical modeling of various linguistic phenomena have shown that additional features give consistent performance improvements. Quite often, improvements are limited by the number of features a system is able to explore. This paper describes a novel progressive training algorithm that selects features from virtually unlimited feature spaces for conditional maximum entr...

This paper presents and evaluates several original techniques for the latent classification of biographic attributes such as gender, age and native language, in diverse genres (conversation transcripts, email) and languages (Arabic, English). First, we present a novel partner-sensitive model for extracting biographic attributes in conversations, given the differences in lexical usage and discou...