A smartphone-based ASR data collection tool for under-resourced languages

Vries, Nic J. de; Davel, Marelie H.; Badenhorst, Jaco; Basson, W. D.; Wet, Febe de; Barnard, Etienne; Waal, Alta de

doi:10.1016/j.specom.2013.07.001

Cited by 74 publications

(44 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…For the Xitsonga experiments, we use data from the NCHLT corpus [37]. Around 12k UTD terms from 24 speakers are discovered from the complete unlabelled training set, with 6.5k true words used for testing.…”

Section: A Datamentioning

confidence: 99%

Unsupervised Feature Learning for Speech Using Correspondence and Siamese Networks

Engelbrecht

Kamper

2020

IEEE Signal Process. Lett.

View full text Add to dashboard Cite

In zero-resource settings where transcribed speech audio is unavailable, unsupervised feature learning is essential for downstream speech processing tasks. Here we compare two recent methods for frame-level acoustic feature learning. For both methods, unsupervised term discovery is used to find pairs of word examples of the same unknown type. Dynamic programming is then used to align the feature frames between each word pair, serving as weak top-down supervision for the two models. For the correspondence autoencoder (CAE), matching frames are presented as input-output pairs. The Triamese network uses a contrastive loss to reduce the distance between frames of the same predicted word type while increasing the distance between negative examples. For the first time, these feature extractors are compared on the same discrimination tasks using the same weak supervision pairs. We find that, on the two datasets considered here, the CAE outperforms the Triamese network. However, we show that a new hybrid correspondence-Triamese approach (CTriamese), consistently outperforms both the CAE and Triamese models in terms of average precision and ABX error rates on both English and Xitsonga evaluation data.

show abstract

Section: A Datamentioning

confidence: 99%

Unsupervised Feature Learning for Speech Using Correspondence and Siamese Networks

Engelbrecht

Kamper

2020

IEEE Signal Process. Lett.

View full text Add to dashboard Cite

show abstract

“…The other set contains carefully uttered, read speech in Xitsonga (2h 29min), a southern African Bantu language. The latter is an excerpt of the NCHLT corpus [22]. All speech segments contain non-overlapping speech of exactly one speaker and are free of non-human noises and pauses.…”

Section: Datamentioning

confidence: 99%

Learning Supervised Feature Transformations on Zero Resources for Improved Acoustic Unit Discovery

Heck

Sakti

Nakamura

2018

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYIn this work we utilize feature transformations that are common in supervised learning without having prior supervision, with the goal to improve Dirichlet process Gaussian mixture model (DPGMM) based acoustic unit discovery. The motivation of using such transformations is to create feature vectors that are more suitable for clustering. The need of labels for these methods makes it difficult to use them in a zero resource setting. To overcome this issue we utilize a first iteration of DPGMM clustering to generate frame based class labels for the target data. The labels serve as basis for learning linear discriminant analysis (LDA), maximum likelihood linear transform (MLLT) and feature-space maximum likelihood linear regression (fMLLR) based feature transformations. The novelty of our approach is the way how we use a traditional acoustic model training pipeline for supervised learning to estimate feature transformations in a zero resource scenario. We show that the learned transformations greatly support the DPGMM sampler in finding better clusters, according to the performance of the DPGMM posteriorgrams on the ABX sound class discriminability task. We also introduce a method for combining posteriorgram outputs of multiple clusterings and demonstrate that such combinations can further improve sound class discriminability.

show abstract

“…As for the second experiment, a 10.5-h and 12-talker subset of the American English Buckeye corpus [24] and a Tsonga dataset [25] containing a total of 4.4 hours of speech from 24 talkers were used similarly to the ZS-2015 challenge (see [16] for details). Following [1], syllable clustering was done in a speaker dependent setting in the second experiment.…”

Section: Data and Pre-processingmentioning

confidence: 99%

Comparison of Non-Parametric Bayesian Mixture Models for Syllable Clustering and Zero-Resource Speech Processing

Seshadri¹,

Remes²,

Räsänen³

2017

Interspeech 2017

View full text Add to dashboard Cite

Zero-resource speech processing (ZS) systems aim to learn structural representations of speech without access to labeled data. A starting point for these systems is the extraction of syllable tokens utilizing the rhythmic structure of a speech signal. Several recent ZS systems have therefore focused on clustering such syllable tokens into linguistically meaningful units. These systems have so far used heuristically set number of clusters, which can, however, be highly dataset dependent and cannot be optimized in actual unsupervised settings. This paper focuses on improving the flexibility of ZS systems using Bayesian non-parametric (BNP) mixture models that are capable of simultaneously learning the cluster models as well as their number based on the properties of the dataset. We also compare different model design choices, namely priors over the weights and the cluster component models, as the impact of these choices is rarely reported in the previous studies. Experiments are conducted using conversational speech from several languages. The models are first evaluated in a separate syllable clustering task and then as a part of a full ZS system in order to examine the potential of BNP methods and illuminate the relative importance of different model design choices.

show abstract

A smartphone-based ASR data collection tool for under-resourced languages

Cited by 74 publications

References 11 publications

Unsupervised Feature Learning for Speech Using Correspondence and Siamese Networks

Unsupervised Feature Learning for Speech Using Correspondence and Siamese Networks

Learning Supervised Feature Transformations on Zero Resources for Improved Acoustic Unit Discovery

Comparison of Non-Parametric Bayesian Mixture Models for Syllable Clustering and Zero-Resource Speech Processing

Contact Info

Product

Resources

About