A Bayesian framework for word segmentation: exploring the effects of context.

By Sharon Goldwater, T. Griffiths, Mark Johnson. Cognition, 2009

493 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Developmental and Educational Psychology

View Paper: https://www.semanticscholar.org/paper/b76761d958c311d0ed8191927ef3fe8d62cf6be6

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

In Developmental and Educational Psychology

Citations in the corpus

Unsupervised Lexicon Discovery from Acoustic Input (2015, background): “…sequences in the input (Cartwright and Brent, 1994; Brent, 1999a; Goldsmith, 2001; Argamon et al., 2004; Goldsmith, 2006; Creutz and Lagus, 2007; Goldwater et al., 2009; Mochihashi et al., 2009; El-sner et al., 2013; Neubig et al., 2012; Heymann et al., 2013; De Marcken, 1996c; De Marcken,…”
A Nonparametric Bayesian Approach to Acoustic Model Discovery (2012, methodology): “Our problem is, in principle, similar to the word segmentation problem discussed in (Goldwater, 2009).”
Unsupervised Learning of Spoken Language with Visual Context (2016, background): “Other approaches have focused on inferring the lexicon of a language from strings of phonemes [10; 11], as well as inferring phone-like and higher level units directly from the speech audio itself [12; 13].”
Bootstrapping language acquisition. (2017, methodology): “Recent probabilistic models of language acquisition (Alishahi & Stevenson, 2008, 2010; Feldman, Griffiths, Goldwater, & Morgan, 2013; Frank, Feldman, & Goldwater, 2014; Goldwater et al., 2009), have demonstrated that Dirichlet Processes can be used to successfully model such infinite distributions, ”
Learning Word-Like Units from Joint Audio-Visual Analysis (2017, background): “categories, and related works aimed to segment and cluster either reference or learned phonemelike tokens into higher-level units (Johnson, 2008; Goldwater et al., 2009; Lee et al., 2015).”
A summary of the 2012 JHU CLSP workshop on zero resource speech technologies and models of early language acquisition (2013, methodology): “Results from the unigram and bigram models are from a single sample after 10k iterations using the hyperparameter values from [19].”
Compositional Imitation Learning: Explaining and executing one task at a time (2018, background): “Unsupervised segmentation and encoding of sequence data is a similarly important problem in natural language or speech processing, e.g., in the context of word or phoneme segmentation [32, 33, 34], or in the segmentation of sequential activity data [35, 36].”
Probabilistic models of cognition: exploring representations and inductive biases. (2010, unknown): “development, from the acquisition of syntax [31, 33, 34] to word segmentation [35] to”
The zero resource speech challenge 2017 (2017, methodology): “The topline was an exhaustive-parsing word segmentation model based on the textual transcriptions (a unigram grammar trained in the adaptor grammar framework: [32]).”
A Tutorial on Bayesian Nonparametric Models (2011, methodology): “BNP models have been fruitfully applied in several other domains, including word segmentation (Goldwater et al., 2009), relational theory acquisition (Kemp et al., 2010) and function learning (Austerweil and Griffiths, 2009b).”
Exploring TTS without T Using Biologically/Psychologically Motivated Neural Network Modules (ZeroSpeech 2020) (2020, background): “A popular approach to such clustering problems in computational linguistics and cognitive science is Bayesian clustering based on the Dirichlet distribution/process [6, 7, 8, 9, 10, 11, 12, 13, 14].”
Discovering latent causes in reinforcement learning (2015, background): “While we have focused on Pavlovian conditioning, the same principles of state discovery, and structure learning more generally, have been invoked across numerous cognitive domains [42–45,18,20,10,46 ,47 ].”
The Zero Resource Speech Challenge 2020: Discovering discrete subword and word units (2020, methodology): “The topline system was an exhaustive-parsing word segmentation model based on the textual transcriptions (a unigram grammar trained in the adaptor grammar framework: [22]).”
A segmental framework for fully-unsupervised large-vocabulary speech recognition (2016, background): “Early work considered full-coverage word segmentation of transcribed phonemic or phonetic symbol sequences (Goldwater et al., 2009; Mochihashi et al., 2009; Neubig et al., 2010; Heymann et al., 2013).”
Unsupervised Latent Tree Induction with Deep Inside-Outside Recursive Auto-Encoders (2019, background): “Learning from Raw Text Unsupervised learning of syntactic structure has been an active research area (Brill et al., 1990), including for un-supervised segmentation (Ando and Lee, 2000; Goldwater et al., 2009; Ponvert et al., 2011) and unsupervised dependency parsing (Spitkovsky et al., 2013).”
A Convolutional Deep Markov Model for Unsupervised Speech Representation Learning (2020, background): “One of the long-standing goals of speech and cognitive scientists is to develop a computational model of language acquisition [1, 2, 3, 4].”
Bayesian Learning of a Tree Substitution Grammar (2009, methodology): “For an excellent introduction to collapsed Gibbs sampling with a DP prior, we refer the reader to Appendix A of Goldwater et al. (2009), which we follow closely here.”
Fully unsupervised small-vocabulary speech recognition using a segmental Bayesian model (2015, methodology): “[12]; their model took transcribed phoneme sequences as input, while our model operates on continuous speech audio.”