Hybrid Dialog State Tracker with ASR Features

Vodolán, Miroslav; Kadlec, Rudolf; Kleindienst, Jan

doi:10.18653/v1/e17-2033

Cited by 20 publications

(30 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To motivate the work presented here, we categorise prior research according to their reliance (or otherwise) on a separate SLU module for interpreting user utterances: 1 Separate SLU Traditional SDS pipelines use Spoken Language Understanding (SLU) decoders to detect slot-value pairs expressed in the Automatic Speech Recognition (ASR) output. The downstream DST model then combines this information with the past dialogue context to update the belief state Wang and Lemon, 2013;Perez, 2016;Sun et al, 2016;Jang et al, 2016;Shi et al, 2016;Dernoncourt et al, 2016;Vodolán et al, 2017). Figure 3: Architecture of the NBT Model.…”

Section: Introductionmentioning

confidence: 99%

Neural Belief Tracker: Data-Driven Dialogue State Tracking

Mrkšić¹,

Séaghdha²,

Wen³

et al. 2017

Proceedings of the 55th Annual Meeting of the Association For Computational Linguistics (Volume 1: Long Papers)

358

252

View full text Add to dashboard Cite

One of the core components of modern spoken dialogue systems is the belief tracker, which estimates the user's goal at every step of the dialogue. However, most current approaches have difficulty scaling to larger, more complex dialogue domains. This is due to their dependency on either: a) Spoken Language Understanding models that require large amounts of annotated training data; or b) hand-crafted lexicons for capturing some of the linguistic variation in users' language. We propose a novel Neural Belief Tracking (NBT) framework which overcomes these problems by building on recent advances in representation learning. NBT models reason over pre-trained word vectors, learning to compose them into distributed representations of user utterances and dialogue context. Our evaluation on two datasets shows that this approach surpasses past limitations, matching the performance of state-of-the-art models which rely on hand-crafted semantic lexicons and outperforming them when such lexicons are not provided.

show abstract

Section: Introductionmentioning

confidence: 99%

Neural Belief Tracker: Data-Driven Dialogue State Tracking

Mrkšić¹,

Séaghdha²,

Wen³

et al. 2017

Proceedings of the 55th Annual Meeting of the Association For Computational Linguistics (Volume 1: Long Papers)

358

252

View full text Add to dashboard Cite

show abstract

“…The Dialogue State Tracking Challenge (DSTC) shared task series formalised the evaluation and provided labelled DST datasets (Henderson et al, 2014a,b;Williams et al, 2016). While a plethora of DST models are available based on, e.g., handcrafted rules or conditional random fields (Lee and Eskenazi, 2013), the recent DST methodology has seen a shift towards neural-network architectures (Henderson et al, 2014c,d;Zilka and Jurcicek, 2015;Mrkšić et al, 2015;Vodolán et al, 2017;Mrkšić et al, 2017a, i.a. ).…”

Section: Downstream Task: Dialogue State Tracking (Dst)mentioning

confidence: 99%

Morph-fitting: Fine-Tuning Word Vector Spaces with Simple Language-Specific Rules

Vulić¹,

Mrkšić²,

Reichart³

et al. 2017

Proceedings of the 55th Annual Meeting of the Association For Computational Linguistics (Volume 1: Long Papers)

View full text Add to dashboard Cite

Morphologically rich languages accentuate two properties of distributional vector space models: 1) the difficulty of inducing accurate representations for lowfrequency word forms; and 2) insensitivity to distinct lexical relations that have similar distributional signatures. These effects are detrimental for language understanding systems, which may infer that inexpensive is a rephrasing for expensive or may not associate acquire with acquires. In this work, we propose a novel morph-fitting procedure which moves past the use of curated semantic lexicons for improving distributional vector spaces. Instead, our method injects morphological constraints generated using simple language-specific rules, pulling inflectional forms of the same word close together and pushing derivational antonyms far apart. In intrinsic evaluation over four languages, we show that our approach: 1) improves low-frequency word estimates; and 2) boosts the semantic quality of the entire word vector collection. Finally, we show that morph-fitted vectors yield large gains in the downstream task of dialogue state tracking, highlighting the importance of morphology for tackling long-tail phenomena in language understanding tasks.

show abstract

“…The current state of the art on the DSTC2 dataset in terms of joint goals accuracy is an ensemble of neural models based on hand-crafted update rules and RNNs (Vodolán et al, 2017). Besides, this model uses a delexicalization mechanism that replaces mentions of slots or values from the DSTC2 ontology by a placeholder to learn valueindependent patterns (Henderson et al, 2014c,b).…”

Section: Comparison To the State Of The Artmentioning

confidence: 99%

Encoding Word Confusion Networks with Recurrent Neural Networks for Dialog State Tracking

Jagfeld

2017

Proceedings of the Workshop on Speech-Centric Natural Language Processing

View full text Add to dashboard Cite

This paper presents our novel method to encode word confusion networks, which can represent a rich hypothesis space of automatic speech recognition systems, via recurrent neural networks. We demonstrate the utility of our approach for the task of dialog state tracking in spoken dialog systems that relies on automatic speech recognition output. Encoding confusion networks outperforms encoding the best hypothesis of the automatic speech recognition in a neural system for dialog state tracking on the well-known second Dialog State Tracking Challenge dataset.

show abstract

Hybrid Dialog State Tracker with ASR Features

Cited by 20 publications

References 20 publications

Neural Belief Tracker: Data-Driven Dialogue State Tracking

Neural Belief Tracker: Data-Driven Dialogue State Tracking

Morph-fitting: Fine-Tuning Word Vector Spaces with Simple Language-Specific Rules

Encoding Word Confusion Networks with Recurrent Neural Networks for Dialog State Tracking

Contact Info

Product

Resources

About