String-based minimum verification error (SB-MVE) training for speech recognition

Rahim, Mazin G.; Lee, Chin‐Hui

doi:10.1006/csla.1997.0026

Cited by 13 publications

(6 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The above models K T and K C can be estimated based on different criteria, such as maximum likelihood (ML), or minimum verification error (MVE), etc. shows the ML-trained models already significantly surpass the conventional UV methods, such as in (Sukkar and Lee, 1996;Sukkar et al, 1997;Rahim et al, 1997;Rahim and Lee, 1997a).…”

Section: In-search Data Selection For Accurate Competing Modelsmentioning

confidence: 96%

“…In , it is found that the minimum classification error (MCE) training, which is originally proposed to reduce recognition errors, can contribute to improving performance of UV. In (Rahim and Lee, 1997a;Sukkar et al, 1997), a GPD-based training algorithm is proposed to achieve minimum verification error (MVE) estimation for utterance verification with respect to optimizing verification HMM parameters. In MVE, the string-level verification errors are approximated by using a sigmoid function embedded with a misverification function, which actually is negative log-likelihood ratio used in verification.…”

Section: As Utterance Verificationmentioning

confidence: 99%

See 1 more Smart Citation

Confidence measures for speech recognition: A survey

Jiang

2005

Speech Communication

308

163

View full text Add to dashboard Cite

In speech recognition, confidence measures (CM) are used to evaluate reliability of recognition results. A good confidence measure can largely benefit speech recognition systems in many practical applications. In this survey, I summarize most research works related to confidence measures which have been done during the past 10-12 years. I will present all these approaches as three major categories, namely CM as a combination of predictor features, CM as a posterior probability, and CM as utterance verification. Then, I also introduce some recent advances in the area. Moreover, I will discuss capabilities and limitations of the current CM techniques and generally comment on todayÕs CM approaches. Based on the discussion, I will conclude the paper with some clues for future works.

show abstract

Section: In-search Data Selection For Accurate Competing Modelsmentioning

confidence: 96%

Section: As Utterance Verificationmentioning

confidence: 99%

Confidence measures for speech recognition: A survey

Jiang

2005

Speech Communication

308

163

View full text Add to dashboard Cite

show abstract

“…misclassification measure), similar to log-likelihood ratio approach in the Bayes decision theory [27]. It was shown that discriminative learning outperforms a binary classification manner in automatic speech recognition and applied in minimum error classification [28] and minimum verification error [29]. The second approach explores the maximal figure-of-merit (MFoM) [30], [31] learning solution, which allows us to approximate the metrics of interest, namely the micro-F1 and equal error rate (EER), with a differentiable function, so that gradient-based optimization algorithms can be applied to learn DNN parameters.…”

Section: Motivationmentioning

confidence: 99%

Maximal Figure-of-Merit Framework to Detect Multi-Label Phonetic Features for Spoken Language Recognition

Kukanov

Trọng

Hautamäki

et al. 2020

IEEE/ACM Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

Bottleneck features (BNFs) generated with a deep neural network (DNN) have proven to boost spoken language recognition accuracy over basic spectral features significantly. However, BNFs are commonly extracted using language-dependent tied-context phone states as learning targets. Moreover, BNFs are less phonetically expressive than the output layer in a DNN, which is usually not used as a speech feature because of its very high dimensionality hindering further post-processing. In this article, we put forth a novel deep learning framework to overcome all of the above issues and evaluate it on the 2017 NIST Language Recognition Evaluation (LRE) challenge. We use manner and place of articulation as speech attributes, which lead to low-dimensional "universal" phonetic features that can be defined across all spoken languages. To model the asynchronous nature of the speech attributes while capturing their intrinsic relationships in a given speech segment, we introduce a new training scheme for deep architectures based on a Maximal Figure of Merit (MFoM) objective. MFoM introduces non-differentiable metrics into the backpropagation-based approach, which is elegantly solved in the proposed framework. The experimental evidence collected on the recent NIST LRE 2017 challenge demonstrates the effectiveness of our solution. In fact, the performance of speech language recognition (SLR) systems based on spectral features is improved for more than 5% absolute Cavg. Finally, the F1 metric can be brought from 77.6% up to 78.1% by combining the conventional baseline phonetic BNFs with the proposed articulatory attribute features.

show abstract

“…It may be desirable to estimate the parameters of a logistic regression model by maximizing F-measure during training. This is analogous, and in a certain sense equivalent, to empirical risk minimization, which has been used successfully in related areas, such as speech recognition (Rahim and Lee, 1997), language modeling (Paciorek and Rosenfeld, 2000), and machine translation (Och, 2003).…”

Section: Introductionmentioning

confidence: 96%

Maximum expected F-measure training of logistic regression models

Jansche

2005

Proceedings of the Conference on Human Language Technology and Empirical Methods in Natural Language Processing - HLT '05

View full text Add to dashboard Cite

We consider the problem of training logistic regression models for binary classification in information extraction and information retrieval tasks. Fitting probabilistic models for use with such tasks should take into account the demands of the taskspecific utility function, in this case the well-known F-measure, which combines recall and precision into a global measure of utility. We develop a training procedure based on empirical risk minimization / utility maximization and evaluate it on a simple extraction task.

show abstract

String-based minimum verification error (SB-MVE) training for speech recognition

Cited by 13 publications

References 18 publications

Confidence measures for speech recognition: A survey

Confidence measures for speech recognition: A survey

Maximal Figure-of-Merit Framework to Detect Multi-Label Phonetic Features for Spoken Language Recognition

Maximum expected F-measure training of logistic regression models

Contact Info

Product

Resources

About