Term-Dependent Confidence Normalisation for Out-of-Vocabulary Spoken Term Detection

Wang, Dong; Tejedor, Javier; King, Simon; Frankel, Joe

doi:10.1007/s11390-012-1228-x

Cited by 20 publications

(42 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We extended this to a discriminative confidence normalization technique (Wang et al, 2009b). This technique provides a general framework which allows any informative factors, or features, to be combined and integrated into an unbiased confidence measure for STD.…”

Section: Motivation and Organization Of This Papermentioning

confidence: 99%

“…This technique provides a general framework which allows any informative factors, or features, to be combined and integrated into an unbiased confidence measure for STD. For example, Wang et al (2009b) used some term-dependent features to compensate for the high diversity among OOV terms, and Tejedor et al (2010) extensively studied various prosodic, lexical and duration-based features.…”

Section: Motivation and Organization Of This Papermentioning

confidence: 99%

“…Zhang and Rudnicky (2001) studied DTs, neural networks and SVMs, and reported that the best performance was given by the SVM. Wang et al (2012) employed MLPs, SVMs and DTs and concluded that all discriminative models provided marginal performance gains and that which model is superior is determined by the properties of the query terms, e.g., whether they are in-vocabulary or out-of-vocabulary.…”

Section: Feature Combinationmentioning

confidence: 99%

See 2 more Smart Citations

Feature analysis for discriminative confidence estimation in spoken term detection

Tejedor

Toledano

Wang

et al. 2014

Computer Speech & Language

Self Cite

View full text Add to dashboard Cite

El acceso a la versión del editor puede requerir la suscripción del recurso Access to the published version may require subscription AbstractDiscriminative confidence based on multi-layer perceptrons (MLPs) and multiple features has shown significant advantage compared to the widely used lattice-based confidence in spoken term detection (STD). Although the MLP-based framework can handle any features derived from a multitude of sources, choosing all possible features may lead to over complex models and hence less generality. In this paper, we design an extensive set of features and analyze their contribution to STD individually and as a group. The main goal is to choose a small set of features that are sufficiently informative while keeping the model simple and generalizable. We employ two established models to conduct the analysis: one is linear regression which targets for the most relevant features and the other is logistic linear regression which targets for the most discriminative features. We find the most informative features are comprised of those derived from diverse sources (ASR decoding, duration and lexical properties) and the two models deliver highly consistent feature ranks. STD experiments on both English and Spanish data demonstrate significant performance gains with the proposed feature sets.

show abstract

Section: Motivation and Organization Of This Papermentioning

confidence: 99%

Section: Motivation and Organization Of This Papermentioning

confidence: 99%

Section: Feature Combinationmentioning

confidence: 99%

See 1 more Smart Citation

Feature analysis for discriminative confidence estimation in spoken term detection

Tejedor

Toledano

Wang

et al. 2014

Computer Speech & Language

Self Cite

View full text Add to dashboard Cite

show abstract

“…We have shown in previous work [46] that a linear remedy can be used to ameliorate the bias problem for the latticebased confidence. With SPM, however, the bias problem is more significant.…”

Section: B Spm and Confidence Biasmentioning

confidence: 99%

“…The term detector was implemented with Lattice2Multigram generously provided to us by the Speech Processing Group, FIT, Brno University of Technology. Term-dependent normalisation [46] was applied in all experiments. The metrics used to evaluate STD performance are ATWV and DET curves; ATWV values with the optimal balance of P miss and P F A are presented as well, denoted by max-ATWV.…”

Section: A Experimental Settingsmentioning

confidence: 99%

Stochastic Pronunciation Modeling for Out-of-Vocabulary Spoken Term Detection

Wang

King

Frankel

2011

IEEE Trans. Audio Speech Lang. Process.

Self Cite

View full text Add to dashboard Cite

Abstract-Spoken term detection (STD) is the name given to the task of searching large amounts of audio for occurrences of spoken terms, which are typically single words or short phrases. One reason that STD is a hard task is that search terms tend to contain a disproportionate number of out-of-vocabulary (OOV) words. The most common approach to STD uses subword units. This, in conjunction with some method for predicting pronunciations of OOVs from their written form, enables the detection of OOV terms but performance is considerably worse than for in-vocabulary terms. This performance differential can be largely attributed to the special properties of OOVs.One such property is the high degree of uncertainty in the pronunciation of OOVs. We present a stochastic pronunciation model (SPM) which explicitly deals with this uncertainty. The key insight is to search for all possible pronunciations when detecting an OOV term, explicitly capturing the uncertainty in pronunciation. This requires a probabilistic model of pronunciation, able to estimate a distribution over all possible pronunciations. We use a joint-multigram model (JMM) for this and compare the JMM-based SPM with the conventional soft match approach. Experiments using speech from the meetings domain demonstrate that the SPM performs better than soft match in most operating regions, especially at low false alarm probabilities. Furthermore, SPM and soft match are found to be complementary: their combination provides further performance gains.

show abstract