Unsupervised Audio Segmentation using Extended Baum-Welch Transformations

Sainath, Tara N.; Kanevsky, Dimitri; Iyengar, G.

doi:10.1109/icassp.2007.366653

Cited by 14 publications

(21 citation statements)

References 3 publications

(6 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In this paper, we expanded on previous work ( [6], [7]), showing that the EBW transformations appears to be a general technique to explain the quality of a model used to represent the data. Specifically, we explored doing BPC recognition using a relative EBW measurement, which we found was able outperform the standard likelihood metric.…”

Section: Discussionmentioning

confidence: 99%

“…In this paper, we continue to expand on previous work ( [6], [7]) and demonstrate that the EBW gradient steepness measure appears to be a general technique to explain the quality of a model used to represent the data. First, we introduce a novel change to our EBW gradient measurement and explain model fit to the data by looking at a relative, rather than absolute, change in the gradient.…”

Section: Introductionmentioning

confidence: 99%

“…In [6] we redefined the likelihood ratio test, typically used for unsupervised segmentation, with this measure of gradient steepness. We showed that our EBW unsupervised audio segmentation method offered improvements over the Bayesian Information Criterion and Cumulative Sum methods.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Broad phonetic class recognition in a Hidden Markov model framework using extended Baum-Welch transformations

Sainath

Kanevsky

Ramabhadran

2007

2007 IEEE Workshop on Automatic Speech Recognition &Amp; Understanding (ASRU)

Self Cite

View full text Add to dashboard Cite

In many pattern recognition tasks, given some input data and a model, a probabilistic likelihood score is often computed to measure how well the model describes the data. Extended Baum-Welch (EBW) transformations are most commonly used as a discriminative technique for estimating parameters of Gaussian mixtures, though recently they have been used to derive a gradient steepness measurement to evaluate the quality of the model to match the distribution of the data. In this paper, we explore applying the EBW gradient steepness metric in the context of Hidden Markov Models (HMMs) for recognition of broad phonetic classes and present a detailed analysis and results on the use of this gradient metric on the TIMIT corpus. We find that our gradient metric is able to outperform the baseline likelihood method, and offers improvements in noisy conditions.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Broad phonetic class recognition in a Hidden Markov model framework using extended Baum-Welch transformations

Sainath

Kanevsky

Ramabhadran

2007

2007 IEEE Workshop on Automatic Speech Recognition &Amp; Understanding (ASRU)

Self Cite

View full text Add to dashboard Cite

show abstract

“…In [5] the likelihood ratio test, typically used for audio segmentation tasks, was redefined with the EBW gradient steepness criteria, while in [6] we explored using EBW for audio classification. In addition, in [7] the EBW metric was used in Hidden Markov Models (HMMs) and showed improvements over the likelihood metric for phonetic recognition.…”

Section: Introductionmentioning

confidence: 99%

Gradient steepness metrics using extended Baum-Welch transformations for universal pattern recognition tasks

Sainath¹,

Kanevsky

Ramabhadran

2008

2008 IEEE International Conference on Acoustics, Speech and Signal Processing

Self Cite

View full text Add to dashboard Cite

In many pattern recognition tasks, given some input data and a family of models, the "best" model is defined as the one which maximizes the likelihood of the data given the model. Extended BaumWelch (EBW) transformations are most commonly used as a discriminative technique for estimating parameters of Gaussian mixtures. In this paper, we use the EBW transformations to derive a novel gradient steepness measurement to find which model best explains the data. We use this gradient measurement to derive a variety of EBW metrics to explain model fit to the data. We apply these EBW metrics to audio segmentation via Hidden Markov Models (HMMs) and show that our gradient steepness measurement is robust across different EBW metrics and model complexities.

show abstract

“…In [SKI07], an unsupervised algorithm for audio segmentation is proposed and applied to the database of meeting-room isolated acoustic events produced in the CHIL project (see Appendix A).…”

Section: Audio Recognition For a Given Environmentmentioning

confidence: 99%

Acoustic Event Detection and Classification

Temko

Nadeu

Macho

et al. 2009

Computers in the Human Interaction Loop

View full text Add to dashboard Cite

The human activity that takes place in meeting-rooms or class-rooms is reflected in a rich variety of acoustic events, either produced by the human body or by objects handled by humans, so the determination of both the identity of sounds and their position in time may help to detect and describe that human activity. Additionally, detection of sounds other than speech may be useful to enhance the robustness of speech technologies like automatic speech recognition.Automatic detection and classification of acoustic events is the objective of this thesis work. It aims at processing the acoustic signals collected by distant microphones in meeting-room or classroom environments to convert them into symbolic descriptions corresponding to a listener's perception of the different sound events that are present in the signals and their sources.First of all, the task of acoustic event classification is faced using Support Vector Machine (SVM) classifiers, which are motivated by the scarcity of training data. A confusion-matrix-based variable-feature-set clustering scheme is developed for the multiclass recognition problem, and tested on the gathered database. With it, a higher classification rate than the GMM-based technique is obtained, arriving to a large relative average error reduction with respect to the best result from the conventional binary tree scheme. Moreover, several ways to extend SVMs to sequence processing are compared, in an attempt to avoid the drawback of SVMs when dealing with audio data, i.e. their restriction to work with fixed-length vectors, observing that the dynamic time warping kernels work well for sounds that show a temporal structure. Furthermore, concepts and tools from the fuzzy theory are used to investigate, first, the importance of and degree of interaction among features, and second, ways to fuse the outputs of several classification systems. The developed AEC systems are Two kinds of databases are used: two databases of isolated acoustic events, and a database of interactive seminars containing a significant number of acoustic events of interest. Our developed systems, which consist of SVM-based classification within a sliding window plus post-processing, were the only submissions not using HMMs, and each of them obtained competitive results in the corresponding evaluation.vi Speech activity detection was also pursued in this thesis since, in fact, it is a -especially important -particular case of acoustic event detection. An enhanced SVM training approach for the speech activity detection task is developed, mainly to cope with the problem of dataset reduction.The resulting SVM-based system is tested with several NIST Rich Transcription (RT) evaluation datasets, and it shows better scores than our GMM-based system, which ranked among the best systems in the RT06 evaluation.Finally, it is worth mentioning a few side outcomes from this thesis work. As it has been carried out in the framework of the CHIL EU project, the author has been responsible for the organization of the above m...

show abstract

Unsupervised Audio Segmentation using Extended Baum-Welch Transformations

Cited by 14 publications

References 3 publications

Broad phonetic class recognition in a Hidden Markov model framework using extended Baum-Welch transformations

Broad phonetic class recognition in a Hidden Markov model framework using extended Baum-Welch transformations

Gradient steepness metrics using extended Baum-Welch transformations for universal pattern recognition tasks

Acoustic Event Detection and Classification

Contact Info

Product

Resources

About