Use of acoustic prior information for confidence measure in ASR (<i>automatic speech recognition</i>) applications

Mengusoglu, Erhan; Ris, Christophe

doi:10.1121/1.1843171

Cited by 4 publications

(5 citation statements)

References 13 publications

(8 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In This result is quite similar to that obtained using the average CM (6). The advantage of this technique, however, is that the threshold has a meaningful interpretation and is easily adjustable according to the application requirements.…”

Section: Normalized Confidence Measuresupporting

confidence: 77%

Confidence measures in multiple pronunciations modeling for speaker verification

BenZeghiba

Bourlard

2004 IEEE International Conference on Acoustics, Speech, and Signal Processing

View full text Add to dashboard Cite

Abstract. This paper investigates the use of multiple pronunciations modeling for UserCustomized Password Speaker Verification (UCP-SV). The main characteristic of the UCP-SV is that the system does not have any a priori knowledge about the password used by the speaker. Our aim is to exploit the information about how the speaker pronounces a password in the decision process. This information is extracted automatically by using a speaker-independent speech recognizer. In this paper, we investigate and compare several techniques. Some of them are based on the combination of confidence scores estimated by different models. In this context, we propose a new confidence measure that uses acoustic information extracted during the speaker enrollment and based on log likelihood ratio measure. These techniques show significant improvement (15.7% relative improvement in terms of equal error rate) compared to a UCP-SV baseline system where the speaker is modeled by only one model (corresponding to one utterance).

show abstract

Section: Normalized Confidence Measuresupporting

confidence: 77%

Confidence measures in multiple pronunciations modeling for speaker verification

BenZeghiba

Bourlard

2004 IEEE International Conference on Acoustics, Speech, and Signal Processing

View full text Add to dashboard Cite

show abstract

“…To avoid substantial variance in the DNN output, we use the confidence measurement [ 34 ] to analyze the output of the DNN-based NC. Based on the confidence measurement score, a threshold is used to determine the classification results.…”

Section: Methodsmentioning

confidence: 99%

Improved Environment-Aware–Based Noise Reduction System for Cochlear Implant Users Based on a Knowledge Transfer Approach: Development and Usability Study

Li¹,

Han²,

Zheng³

et al. 2021

J Med Internet Res

View full text Add to dashboard Cite

Background Cochlear implant technology is a well-known approach to help deaf individuals hear speech again and can improve speech intelligibility in quiet conditions; however, it still has room for improvement in noisy conditions. More recently, it has been proven that deep learning–based noise reduction, such as noise classification and deep denoising autoencoder (NC+DDAE), can benefit the intelligibility performance of patients with cochlear implants compared to classical noise reduction algorithms. Objective Following the successful implementation of the NC+DDAE model in our previous study, this study aimed to propose an advanced noise reduction system using knowledge transfer technology, called NC+DDAE_T; examine the proposed NC+DDAE_T noise reduction system using objective evaluations and subjective listening tests; and investigate which layer substitution of the knowledge transfer technology in the NC+DDAE_T noise reduction system provides the best outcome. Methods The knowledge transfer technology was adopted to reduce the number of parameters of the NC+DDAE_T compared with the NC+DDAE. We investigated which layer should be substituted using short-time objective intelligibility and perceptual evaluation of speech quality scores as well as t-distributed stochastic neighbor embedding to visualize the features in each model layer. Moreover, we enrolled 10 cochlear implant users for listening tests to evaluate the benefits of the newly developed NC+DDAE_T. Results The experimental results showed that substituting the middle layer (ie, the second layer in this study) of the noise-independent DDAE (NI-DDAE) model achieved the best performance gain regarding short-time objective intelligibility and perceptual evaluation of speech quality scores. Therefore, the parameters of layer 3 in the NI-DDAE were chosen to be replaced, thereby establishing the NC+DDAE_T. Both objective and listening test results showed that the proposed NC+DDAE_T noise reduction system achieved similar performances compared with the previous NC+DDAE in several noisy test conditions. However, the proposed NC+DDAE_T only required a quarter of the number of parameters compared to the NC+DDAE. Conclusions This study demonstrated that knowledge transfer technology can help reduce the number of parameters in an NC+DDAE while keeping similar performance rates. This suggests that the proposed NC+DDAE_T model may reduce the implementation costs of this noise reduction system and provide more benefits for cochlear implant users.

show abstract

“…Typical spoken language systems consist of two main sub-systems: an ASR (automatic speech recognition) front-end, which generates a recognition hypothesis (or N-best list of recognition hypotheses) for each input utterance, and an NLP (natural language processing) back-end, which performs semantic understanding, dialogue management, and response generation. While conventional approaches [8]- [15] generate confidence measures based on the information obtained during decoding in the ASR front-end, this paper focuses on the incorporation of "high-level " knowledge sources from the back-end system.…”

Section: Proposed Confidence Scoring Frameworkmentioning

confidence: 99%

“…Explicit model-based schemes [10]- [12] compare the candidate model to a competing model (an anti-model, background model or set of cohort models) via a likelihood ratio test. Posterior probability-based approaches, including [13]- [15], estimate the posterior probability of a recognized entity (word or utterance) considering all competing hypotheses (typically in an N-best list or word graph).…”

Section: Introductionmentioning

confidence: 99%

Verification of Speech Recognition Results Incorporating In-domain Confidence and Discourse Coherence Measures

Lane

Kawahara

2006

IEICE Transactions on Information and Systems

View full text Add to dashboard Cite

Conventional confidence measures for assessing the reliability of ASR (automatic speech recognition) output are typically derived from "low-level" information which is obtained during speech recognition decoding. In contrast to these approaches, we propose a novel utterance verification framework which incorporates "high-level" knowledge sources. Specifically, we investigate two measures: in-domain confidence, the degree of match between the input utterance and the application domain of the back-end system, and discourse coherence, the consistency between consecutive utterances in a dialogue session. A joint confidence score is generated by combining these two measures with an orthodox measure based on GPP (generalized posterior probability). The proposed framework was evaluated on an utterance verification task for spontaneous dialogue performed via a (English/Japanese) speech-to-speech translation system. Incorporating the two proposed measures significantly improved utterance verification accuracy compared to using GPP alone, realizing reductions in CER (confidence error-rate) of 11.4% and 8.1% for the English and Japanese sides, respectively. When negligible ASR errors (that do not significantly affect translation) were ignored, further improvement was achieved for the English side, realizing a reduction in CER of up to 14.6% compared to the GPP case.

show abstract

Use of acoustic prior information for confidence measure in ASR (automatic speech recognition) applications

Cited by 4 publications

References 13 publications

Confidence measures in multiple pronunciations modeling for speaker verification

Confidence measures in multiple pronunciations modeling for speaker verification

Improved Environment-Aware–Based Noise Reduction System for Cochlear Implant Users Based on a Knowledge Transfer Approach: Development and Usability Study

Verification of Speech Recognition Results Incorporating In-domain Confidence and Discourse Coherence Measures

Contact Info

Product

Resources

About