Parallel inference of dirichlet process Gaussian mixture models for unsupervised acoustic modeling: a feasibility study

By Hongjie Chen, Cheung Chi Leung, Lei Xie, Bin Ma, Haizhou Li. 2015

76 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Probabilistic & Bayesian Methods, 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 Probabilistic & Bayesian Methods

In Speech & Audio

In Signal Processing

Citations in the corpus

VQVAE Unsupervised Unit Discovery and Multi-scale Code2Spec Inverter for Zerospeech Challenge 2019 (2019, methodology): “The top performances in discovering speech representation in ZeroSpeech 2015 and 2017 are dominated by a Bayesian non-parametric approach with unsupervised cluster speech features using a Dirichlet process Gaussian mixture model (DPGMM) [4, 5].”
Unsupervised Speech Representation Learning Using WaveNet Autoencoders (2019, methodology): “DP-GMM [66] imposes a Dirichlet Process prior over a Gaussian Mixture Model.”
The zero resource speech challenge 2017 (2017, methodology): “The ﬁrst strategy consists in performing a bottom-up frame-level clustering, following up on the success of [20] with this simple strategy.”
UWSpeech: Speech to Speech Translation for Unwritten Languages (2020, background): “2017), Gaussian mixture model clustering (Chen et al. 2015), tree-based clustering (Muthukumar and Black 2014), binarization with straight-through estimation (Eloff et al.”
A segmental framework for fully-unsupervised large-vocabulary speech recognition (2016, methodology): “Approaches include those using bottom-up trained Gaussian mixture models (GMMs) to produce frame-level posteriorgrams (Zhang and Glass, 2010; Chen et al., 2015), using unsupervised hidden Markov models (HMMs) to obtain discrete categorical output in terms of discovered subword units (Varadarajan et ”
The Zero Resource Speech Challenge 2015: Proposed Approaches and Results (2016, methodology): “The systems based around modeling the acoustic space in an unsupervised manner [3, 4, 6] provide insight into the effectiveness of these methods that could be transferred to supervised methods.”
Optimizing DPGMM Clustering in Zero Resource Setting Based on Functional Load (2018, methodology): “For comparison, we use the same parameter setting as [2, 3] for DPGMM sampling; we use officially provided voice activity detection segmentations (2.”
Multilingual bottle-neck feature learning from untranscribed speech (2017, methodology): “All the hyperparameters for the inference of all DPGMMs are set following our previous work [8] and are listed in Table 2, where mean(X ) and cov(X ) is the mean vector and covariance matrix of the input feature vectors, X .”
Speaker Invariant Feature Extraction for Zero-Resource Languages with Adversarial Learning (2018, methodology): “In [6], the sub-word units are automatically generated in nonparametric manner on the target language.”
Feature optimized DPGMM clustering for unsupervised subword modeling: A contribution to zerospeech 2017 (2017, methodology): “[6] – performing best in ZeroSpeech 2015 – take a Bayesian non-parametric approach and cluster MFCCs with their derivatives by inferring a Dirichlet process Gaussian mixture model (DPGMM).”
ABX-Discriminability Measures and Applications. (Mesures de discriminabilité ABX et applications) (2016, background): “9% ABX error rate in the across-speaker task on the Buckeye corpus) were largely above the raw acoustic features baseline, but remained inferior to the results with DPGMM from [106, 107] and those obtained with NN trained based on top-down information from STD [101–103].”
Combining Adversarial Training and Disentangled Speech Representation for Robust Zero-Resource Subword Modeling (2019, methodology): “The reconstructed features are fed as inputs to Dirichlet process Gaussian mixture model (DPGMM) [19] for frame clustering, as was done in [4].”
A deep scattering spectrum — Deep Siamese network pipeline for unsupervised acoustic modeling (2016, methodology): “The table further shows that ScatABnet is competitive with the state of the art system of [20], in one case, across-speaker for Xitsonga, producing the lowest ABX error.”
An empirical evaluation of zero resource acoustic unit discovery (2017, unknown): “Past studies in [8, 9] demonstrated the effectiveness of posterior features based on automatically derived acoustic structures by spoken term detection and phoneme discrimination tasks, while being limited to frame-level clustering and loss of phonetic temporal information.”
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;…”