Minimum Bayes-risk automatic speech recognition

By Vaibhava Goel, William J. Byrne. 2000

245 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

An Actor-Critic Algorithm for Sequence Prediction (2016, neutral): “Another popular approach is to replace the domain over which the task score expectation is deﬁned with a small subset of it, as is done in Minimum (Bayes) Risk Training (Goel & Byrne, 2000; Shen et al., 2015; Och, 2003).”
Determinantal Point Processes for Machine Learning (2012, methodology): “First proposed by Goel and Byrne [59] for automatic speech recognition, MBR decoding has also been used successfully for word alignment and machine translation [86, 87].”
Minimum Bayes-Risk Word Alignments of Bilingual Texts (2002, methodology): “We have applied Minimum Bayes-Risk (MBR) procedures developed for automatic speech recognition (Goel and Byrne, 2000) to word alignment of bitexts.”
Minimum Bayes-Risk Decoding for Statistical Machine Translation (2004, methodology): “We apply the Minimum Bayes-Risk (MBR) techniques developed for automatic speech recognition (Goel and Byrne, 2000) and bitext word alignment for statistical MT (Kumar and Byrne, 2002), to the problem of building automatic MT systems tuned for specific metrics.”
Is MAP Decoding All You Need? The Inadequacy of the Mode in Neural Machine Translation (2020, methodology): “This decision rule, known as MBR decoding in the NLP literature (Goel and Byrne, 2000), is especially suited where we trust a model in expectation but not its mode in particular (Smith, 2011,”
A Fully Differentiable Beam Search Decoder (2019, methodology): “…2006; Graves and Jaitly, 2014) and the Auto Segmentation (ASG) criterion (Collobert et al., 2016), as well as Minimum Bayes Risk (MBR and sMBR) (Goel and Byrne, 2000; Gibson and Hain, 2006; Sak et al., 2015; Prabhavalkar et al., 2018) and the Maximum Mutual Information (MMI) criterion (Bahl et…”
The Edit Distance Transducer in Action: The University of Cambridge English-German System at WMT16 (2016, background): “MBR also does not enforce exact agreement between systems as it distinguishes between the hypothesis space and the evidence space (Goel and Byrne, 2000; Tromble et al., 2008).”
A Study of Reinforcement Learning for Neural Machine Translation (2018, background): “Shen et al. (2016) proposes to adopt minimum risk training (Goel and Byrne, 2000) 2017) presents a comparative study of several classical structural prediction losses for NMT model, which also includes sequence-level loss but not exactly the same as RL.”
Natural Language to Code Translation with Execution (2022, methodology): “In machine translation, MBR decoding is usually implemented by reranking candidates (Goel and [CODE] with natural language descriptions and corresponding code snippets respectively.”
Finding consensus in speech recognition: word error minimization and other applications of confusion networks (2000, methodology): “An alternative approach recently developed by Goel & Byrne (2000) takes a diﬀerent approach and searches the lattice directly using the original word error cost function.”
Machine Learning Paradigms for Speech Recognition: An Overview (2013, methodology): “One major school of such models are referred to as Bayesian Mininum Risk (BMR) classifiers [66]–[68]:”
Generalized Minimum Bayes Risk System Combination (2011, methodology): “MBR arose in Bayes decision theory (Duda et al., 2000) and has since been applied to speech recognition (Goel and Byrne, 2000) and machine translation (Kumar and Byrne, 2004).”