An auto-encoder based approach to unsupervised learning of subword units

By Leonardo Badino, Claudia Canevari, Luciano Fadiga, Giorgio Metta. 2014

83 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Natural Language Processing, Representation & Self-Supervised Learning

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 Representation & Self-Supervised Learning

Citations in the corpus

Unsupervised acoustic unit discovery for speech synthesis using discrete latent-variable neural networks (2019, methodology): “AEs are well-suited to our task since compression can be directly controlled through the latent layer dimensionality [14,25,26].”
The Zero Resource Speech Challenge 2019: TTS without T (2019, methodology): “On the acoustic unit discovery side, several methods have been used (binarized autoencoders [7], binarized siamese networks [8]; a variety of speaker normalization techniques have been used to improve the categories [9], among other techniques).”
A segmental framework for fully-unsupervised large-vocabulary speech recognition (2016, methodology): “The model signiﬁcantly outperformed the top-down GMM-UBM [32] and stacked AEs [28,29] in an intrinsic evaluation: isolated word discrimination.”
Data Augmenting Contrastive Learning of Speech Representations in the Time Domain (2020, background): “The reconstructed representation can be the raw audio [2, 14, 15], or higher level speech features like MFCC, PLP [16, 17].”
An embedded segmental K-means model for unsupervised segmentation and clustering of speech (2017, background): “Several zero-resource tasks have been studied, including acoustic unit discovery [4–6], unsupervised representation learning [7–9], query-by-example search [10, 11] and topic modelling [12, 13].”
Parallel inference of dirichlet process Gaussian mixture models for unsupervised acoustic modeling: a feasibility study (2015, background): “In additional to posteriorgrams, there are works on frame-based embedding representation motivated by manifold learning [20–22] and deep learning [10,11,23].”
Unsupervised End-to-End Learning of Discrete Linguistic Units for Voice Conversion (2019, background): “Previous works [14][15][16] also attempt to learn discrete representations from audio, however they did not apply the learned units for VC.”
A comparison of neural network methods for unsupervised representation learning on the zero resource speech challenge (2015, unknown): “[15, 27].”
Discovering discrete subword units with binarized autoencoders and hidden-Markov-model encoders (2015, methodology): “We consider this as an interesting result as in [10] more “compact” subword transcriptions always resulted in poorer performance in a word classification task.”
DeepEar: robust smartphone audio sensing in unconstrained acoustic environments using deep learning (2015, background): “Such success has spawned a rich set of audio-focused deep learning techniques and algorithms [23, 33, 10].”
Fully unsupervised small-vocabulary speech recognition using a segmental Bayesian model (2015, background): “Recent studies in the speech processing community have applied unsupervised techniques to tasks such as phonetic discovery [1, 2, 3, 4], lexical discovery [5, 6, 7], spoken document retrieval [8] and query-by-example search [9, 10].”
A Brief Overview of Unsupervised Neural Speech Representation Learning (2022, methodology): “However, sometimes local smoothness is explicitly used to define the task (Badino et al. 2014; Jati and Georgiou 2017, 2019).”
Self-Supervised Speech Representation Learning: A Review (2022, background): “In addition to the approaches discussed above, several other works on speech representation learning take inspiration from the AE framework [86], [87], [88], [89], [90], [91].”
Deep convolutional acoustic word embeddings using word-pair side information (2015, methodology): “Such paired supervision has been used for several problems and domains, including phonetic discovery [12, 21], semantic word embeddings [22–24] and vision applications [25].”
The zero resource speech challenge 2015 (2015, methodology): “Yet other approaches use Deep Neural Net (DNN) architectures with unsupervised or weakly supervised loss functions to learn phone-level embeddings [1, 6, 19].”