Exemplar-Based Processing for Speech Recognition: An Overview

Sainath, Tara N.; Ramabhadran, Bhuvana; Nahamoo, D.; Kanevsky, Dimitri; Compernolle, Dirk Van; Demuynck, Kris; Gemmeke, Jort F.; Bellegarda, Jerome R.; Sundaram, Shiva

doi:10.1109/msp.2012.2208663

Cited by 39 publications

(31 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Sparse codes are also known to be analogous to the coding mechanism in neural sensory system (see [97] and references therein). Interestingly, mathematical analogy between sparse representation of data and dimension reduction using matrix factorization (used in the bag-of-units representation by [93]) has also been observed in the context of speech recognition (the reader is referred to [102] for an overview). Starting from perceptual features, retrieval techniques using these representations can therefore be used to further render higher level sensory processes in the auditory system.…”

Section: A Semantic Audio Retrievalmentioning

confidence: 90%

An Overview on Perceptually Motivated Audio Indexing and Classification

Richard

Sundaram²,

Narayanan

2013

Proc. IEEE

Self Cite

View full text Add to dashboard Cite

An audio indexing system aims at describing audio content by identifying, labeling or categorizing different acoustic events. Since the resulting audio classification and indexing is meant for direct human consumption, it is highly desirable that it produces perceptually relevant results. This can be obtained by integrating specific knowledge of the human auditory system in the design process to various extent. In this paper, we highlight some of the important concepts used in audio classification and indexing that are perceptually motivated or that exploit some principles of perception. In particular, we discuss several different strategies to integrate human perception including 1) the use of generic audition models, 2) the use of perceptually-relevant features for the analysis stage that are perceptually justified either as a component of a hearing model or as being correlated with a perceptual dimension of sound similarity, and 3) the involvement of the user in the audio indexing or classification task. In the paper, we also illustrate some of the recent trends in semantic audio retrieval that approximate higher level perceptual processing and cognitive aspects of human audio recognition capabilities including affect-based audio retrieval.

show abstract

Section: A Semantic Audio Retrievalmentioning

confidence: 90%

An Overview on Perceptually Motivated Audio Indexing and Classification

Richard

Sundaram²,

Narayanan

2013

Proc. IEEE

Self Cite

View full text Add to dashboard Cite

show abstract

“…It can also be viewed as an instance of exemplar-based technique [17] at the speaker level. Most of the previous attempts used some form of approaximation of SI and SD models to select the neighbours based on the acoustic match.…”

Section: Related Workmentioning

confidence: 99%

Neighbour selection and adaptation for rapid speaker-dependent ASR

Nallasamy¹,

Fuhs²,

Woszczyna³

et al. 2013

2013 IEEE Workshop on Automatic Speech Recognition and Understanding

View full text Add to dashboard Cite

Speaker dependent (SD) ASR systems have significantly lower word error rates (WER) compared to speaker independent (SI) systems. However, SD systems require sufficient training data from the target speaker, which is impractical to collect in a short time. We present a technique for training SD models using just few minutes of speaker's data. We compensate for the lack of adequate speaker-specific data by selecting neighbours from a database of existing speakers who are acoustically close to the target speaker. These neighbours provide ample training data, which is used to adapt the SI model to obtain an initial SD model for the new speaker with significantly lower WER. We evaluate various neighbour selection algorithms on a large-scale medical transcription task and report significant reduction in WER using only 5 mins of speaker-specific data. We conduct a detailed analysis of various factors such as gender and accent in the neighbour selection. Finally, we study neighbour selection and adaptation in the context of discriminative objective function.

show abstract

“…The trade-off is that such a system will have a huge space and time complexity. However, with the ever increasing amount of training data, as well as the growing computational and memory resources, the potential of exemplar-based approaches is currently being explored extensively (Sainath et al, 2012(Sainath et al, , 2011Gemmeke et al, 2011;De Wachter et al, 2007).…”

Section: Introductionmentioning

confidence: 99%

Sparse modeling of neural network posterior probabilities for exemplar-based speech recognition

Dighe

Asaei

Bourlard

2016

Speech Communication

View full text Add to dashboard Cite

In this paper, a compressive sensing (CS) perspective to exemplar-based speech processing is proposed. Relying on an analytical relationship between CS formulation and statistical speech recognition (Hidden Markov Models -HMM), the automatic speech recognition (ASR) problem is cast as recovery of highdimensional sparse word representation from the observed low-dimensional acoustic features. The acoustic features are exemplars obtained from (deep) neural network sub-word conditional posterior probabilities. Low-dimensional word manifolds are learned using these sub-word posterior exemplars and exploited to construct a linguistic dictionary for sparse representation of word posteriors. Dictionary learning has been found to be a principled way to alleviate the need of having huge collection of exemplars as required in conventional exemplar-based approaches, while still improving the performance. Context appending and collaborative hierarchical sparsity are used to exploit the sequential and group structure underlying word sparse representation. This formulation leads to a posterior-based sparse modeling approach to speech recognition. The potential of the proposed approach is demonstrated on isolated word (Phonebook corpus) and continuous speech (Numbers corpus) recognition tasks.

show abstract

Exemplar-Based Processing for Speech Recognition: An Overview

Cited by 39 publications

References 41 publications

An Overview on Perceptually Motivated Audio Indexing and Classification

An Overview on Perceptually Motivated Audio Indexing and Classification

Neighbour selection and adaptation for rapid speaker-dependent ASR

Sparse modeling of neural network posterior probabilities for exemplar-based speech recognition

Contact Info

Product

Resources

About