Unsupervised Lexicon Discovery from Acoustic Input

By Chia-ying Lee, Timothy J. O'Donnell, James R. Glass. 2015

92 citations recorded by the source corpus.

We present a model of unsupervised phonological lexicon discovery—the problem of simultaneously learning phoneme-like and word-like units from acoustic input. Our model builds on earlier models of unsupervised phone-like unit discovery from acoustic data (Lee and Glass, 2012), and unsupervised symbolic lexicon discovery using the Adaptor Grammar framework (Johnson et al., 2006), integrating these earlier approaches using a probabilistic model of phonological variation. We show that the model is competitive with state-of-the-art spoken term discovery systems, and present analyses exploring the model’s behavior and the kinds of linguistic structures it learns.

Subfields: Artificial Intelligence, Natural Language Processing, Speech & Audio, Signal Processing

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 Natural Language Processing

In Speech & Audio

In Signal Processing

Citations in the corpus

Unsupervised Speech Recognition (2021, methodology): “…been segmented and clustered in a large body of prior work (Varadarajan et al., 2008; Zhang and Glass, 2009; Gish et al., 2009; Lee and Glass, 2012; Lee et al., 2015; Ondel et al., 2016), including Bayesian methods (Kamper et al., 2017a) and more efficient approximations thereof (Kamper et al.,…”
Learning Hierarchical Discrete Linguistic Units from Visually-Grounded Speech (2019, background): “A different line of work employs probabilistic graphical models to jointly segment and cluster the speech signal (Varadarajan et al., 2008; Zhang & Glass, 2009; Gish et al., 2009; Lee & Glass, 2012; Siu et al., 2014; Lee et al., 2015; Ondel et al., 2016; Kamper et al., 2016; 2017a).”
Unsupervised acoustic unit discovery for speech synthesis using discrete latent-variable neural networks (2019, methodology): “Here we consider discrete representation learning, also referred to as ‘acoustic unit dis-covery’ [12, 13] or ‘discrete subword learning’ [14].”
A segmental framework for fully-unsupervised large-vocabulary speech recognition (2016, methodology): “Furthermore, in contrast to [20–22], our model operates directly at the whole-word level instead of having both word and subword models.”
Unsupervised Cross-Modal Alignment of Speech and Text Embedding Spaces (2018, background): “This has prompted several studies on full-coverage approaches, where the entire speech input is segmented into word-like units [31, 32, 33, 34].”
An embedded segmental K-means model for unsupervised segmentation and clustering of speech (2017, neutral): “These two models are representative of two methodological extremes often seen in zero-resource systems: either probabilistic Bayesian models with convergence guarantees are used [6, 19, 23, 25–27], or heuristic techniques are used in pipeline approaches [18,24].”
Learning Word-Like Units from Joint Audio-Visual Analysis (2017, background): “…Ondel et al., 2016) employed a generative model to cluster acoustic segments into phoneme-like categories, and related works aimed to segment and cluster either reference or learned phoneme-like tokens into word-like and higher-level units (Johnson, 2008; Goldwater et al., 2009; Lee et al., 2015).”
Unsupervised Word Segmentation and Lexicon Discovery Using Acoustic Word Embeddings (2016, methodology): “[12], [23] developed a non-parametric hierarchical”
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].”
Semantic Speech Retrieval With a Visually Grounded Model of Untranscribed Speech (2017, unknown): “ery [39], and unsupervised segmentation and clustering [40], [41].”
Visually Grounded Learning of Keyword Prediction from Untranscribed Speech (2017, background): “Here we consider the problem of grounding un-labelled speech when paired with images.”
Vector-quantized neural networks for acoustic unit discovery in the ZeroSpeech 2020 challenge (2020, methodology): “Early approaches to acoustic unit discovery typically combined clustering methods with hidden Markov models [15–19].”
Multilingual bottle-neck feature learning from untranscribed speech (2017, unknown): “Recently there is an increasing research interest in unsupervised speech processing, which usually involves unsupervised discovery of linguistic units [1, 2, 3], and the derived techniques have been used in different applications, such as retrieval of spoken queries in a speech database [4, 5, 6] an”
Towards Visually Grounded Sub-word Speech Unit Discovery (2019, background): “Other works have jointly learned subword as well as word-level units in a unified framework [8, 9].”