A comparison of neural network methods for unsupervised representation learning on the zero resource speech challenge

By Daniel Renshaw, Herman Kamper, Aren Jansen, Sharon Goldwater, †. Cstr. 2015

103 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Representation & Self-Supervised Learning, 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 Representation & Self-Supervised Learning

In Speech & Audio

In Signal Processing

Citations in the corpus

Unsupervised acoustic unit discovery for speech synthesis using discrete latent-variable neural networks (2019, neutral): “As part of the 2015 and 2017 Zero Resource Speech Challenges (ZRSC), several unsupervised methods were proposed for learning continuous acoustic representations [6–11].”
The zero resource speech challenge 2017 (2017, methodology): “Speciﬁcally, following [28], a stacked autoencoder is trained on DTW-aligned frame pairs from instances of motifs paired by the STD system.”
A segmental framework for fully-unsupervised large-vocabulary speech recognition (2016, methodology): “Although it is common in NN speech systems to use nine or eleven sliding frames as input, we use single-frame MFCCs with ﬁrst and second order derivatives (39-dimensional), as also done in [14,26].”
Jointly Discovering Visual Objects and Spoken Words from Raw Sensory Input (2018, methodology): “Neural networks have thus far been mostly utilized in this realm for learning frame-level acoustic features (Kamper et al. 2015; Renshaw et al. 2015; Thiolliere et al. 2015; Zhang et al. 2012).”
The Zero Resource Speech Challenge 2015: Proposed Approaches and Results (2016, methodology): “Or the feedback from Track 2 into Track 1 as was shown in several systems [9, 10] could be applied to other feature extraction systems.”
An embedded segmental K-means model for unsupervised segmentation and clustering of speech (2017, background): “com/kamperh/recipe_zs2017_track2 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].”
Analysis of Audio-Visual Features for Unsupervised Speech Recognition (2017, background): “Kamper et al. ex-periment with both MFCC features and correspondence autoen-coder (CAE) features [11]; their results suggest the potential of”
Towards Bilingual Lexicon Discovery From Visually Grounded Speech Audio (2019, methodology): “Deep learning has been used to learn robust feature representation for speech over varying speaker and background characteristics such as in [13, 14, 15, 16].”
Measuring the perceptual availability of phonological features during language acquisition using unsupervised binary stochastic autoencoders (2019, background): “Participating systems (Badino et al., 2015; Renshaw et al., 2015; Agenbag and Niesler, 2015; Chen et al., 2015; Baljekar et al., 2015; Räsänen et al., 2015; Lyzinski et al., 2015; Zeghidour et al., 2016; Heck et al., 2016; Srivastava and Shrivastava, 2016; Kamper et al., 2017b; Chen et al., 2017;…”
A Brief Overview of Unsupervised Neural Speech Representation Learning (2022, background): “Prior to this work, Kamper et al. (2015) and Renshaw et al. (2015) introduced the correspondence autoencoder.”
ABX-Discriminability Measures and Applications. (Mesures de discriminabilité ABX et applications) (2016, methodology): “9% AP for the Correspondence Autoencoder method from [102] and 54.”
A deep scattering spectrum — Deep Siamese network pipeline for unsupervised acoustic modeling (2016, background): “In previous work, several architectures have been explored to learn an acoustic model in an unsupervised or weakly-supervised way [2, 3, 4, 5].”
Extracting bottleneck features and word-like pairs from untranscribed speech for feature representation (2017, methodology): “The matching frame pairs were used twice as in [20] when fine-tuning the DNN.”
Deep convolutional acoustic word embeddings using word-pair side information (2015, methodology): “If the labels W train for the training set Y train are not known, a weaker form of supervision that has also been used [11–13, 19] is the To train these networks it is sometimes assumed that pairs not in S train belong to different types; we also make this assumption here.”
Deep learning methods for unsupervised acoustic modeling — Leap submission to ZeroSpeech challenge 2017 (2017, methodology): “Recently, several approaches have been proposed for this task using various feature representations [7] and neural network models [8, 9].”
Combining Adversarial Training and Disentangled Speech Representation for Robust Zero-Resource Subword Modeling (2019, background): “In the previous two challenges, several unsupervised representation learning approaches were proposed for comparison, such as cluster posteriorgrams (PGs) [4–6], DNN bottleneck features [7,8], autoencoders (AEs) [9,10], variational AEs (VAEs) [11, 12] and siamese networks [13–15].”
Optimizing DPGMM Clustering in Zero Resource Setting Based on Functional Load (2018, methodology): “In addition to posteriorgram representation, we can also reduce the dimension of framebased embedding representation using auto-encoder [6, 7].”