Keyword Spotting of Arbitrary Words Using Minimal Speech Resources

Garcia, Angela; Gish, H.

doi:10.1109/icassp.2006.1660179

Cited by 42 publications

(34 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The more recent work of Novotney et al showed the potential for leveraging small amounts of annotated data to train initial models, which can then be used to automatically annotate much larger quantities of unannotated data [5]. The approach by Gish et al requires even less supervision by creating initial acoustic models using unsupervised techniques on unannotated audio data [13]. This latter work is most similar to our ongoing research directed towards rapid portability of ASR technology to languages with limited linguistic resources.…”

Section: Introductionmentioning

confidence: 99%

Unsupervised spoken keyword spotting via segmental DTW on Gaussian posteriorgrams

Zhang¹,

Glass²

2009

2009 IEEE Workshop on Automatic Speech Recognition &Amp; Understanding

289

242

View full text Add to dashboard Cite

Abstract-In this paper, we present an unsupervised learning framework to address the problem of detecting spoken keywords. Without any transcription information, a Gaussian Mixture Model is trained to label speech frames with a Gaussian posteriorgram. Given one or more spoken examples of a keyword, we use segmental dynamic time warping to compare the Gaussian posteriorgrams between keyword samples and test utterances. The keyword detection result is then obtained by ranking the distortion scores of all the test utterances. We examine the TIMIT corpus as a development set to tune the parameters in our system, and the MIT Lecture corpus for more substantial evaluation. The results demonstrate the viability and effectiveness of our unsupervised learning framework on the keyword spotting task.

show abstract

Section: Introductionmentioning

confidence: 99%

Unsupervised spoken keyword spotting via segmental DTW on Gaussian posteriorgrams

Zhang¹,

Glass²

2009

2009 IEEE Workshop on Automatic Speech Recognition &Amp; Understanding

289

242

View full text Add to dashboard Cite

show abstract

“…One of the disadvantages of this approach, however, is poor generalization to arbitrary languages, (or more general audio), since it typically requires a trained speech recognizer. Thus, for under-resourced languages, there is a time/cost issue to obtain enough annotated data to build a recognizer with acceptable recognition performance [2,3].…”

Section: Introductionmentioning

confidence: 99%

Resource configurable spoken query detection using Deep Boltzmann Machines

Zhang¹,

Salakhutdinov

Chang³

et al. 2012

2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

In this paper we present a spoken query detection method based on posteriorgrams generated from Deep Boltzmann Machines (DBMs). The proposed method can be deployed in both semi-supervised and unsupervised training scenarios. The DBM-based posteriorgrams were evaluated on a series of keyword spotting tasks using the TIMIT speech corpus. In unsupervised training conditions, the DBM-approach improved upon our previous best unsupervised keyword detection performance using Gaussian mixture model-based posteriorgrams by over 10%. When limited amounts of labeled data were incorporated into training, the DBM-approach required less than one third of the annotated data in order to achieve a comparable performance of a system that used all of the annotated data for training.

show abstract

“…Converting all matched fragment pairs to a graph. Each numbered node corresponds to a temporal local maximum in fragment similarity in a particular utterance (e.g., [1][2][3][4][5]. Each matching fragment is represented by a connection between two nodes in the graph (e.g., 1-4, 2-4, 3-5).…”

Section: Segmental Dtwmentioning

confidence: 99%

“…Annotation of speech corpora is currently a very time consuming and expensive endeavor and is a limiting factor in how quickly speech recognizers can be created for new problem areas and languages. Given the relative ease of creating and storing large quantities of audio-visual speech material these days, methods that can process vast quantities of unannotated data to enable keyword search [1], audio summarization etc. could be quite useful.…”

Section: Introductionmentioning

confidence: 99%

Towards multi-speaker unsupervised speech pattern discovery

Zhang¹,

Glass²

2010

2010 IEEE International Conference on Acoustics, Speech and Signal Processing

View full text Add to dashboard Cite

In this paper, we explore the use of a Gaussian posteriorgram based representation for unsupervised discovery of speech patterns. Compared with our previous work, the new approach provides significant improvement towards speaker independence. The framework consists of three main procedures: a Gaussian posteriorgram generation procedure which learns an unsupervised Gaussian mixture model and labels each speech frame with a Gaussian posteriorgram representation; a segmental dynamic time warping procedure which locates pairs of similar sequences of Gaussian posteriorgram vectors; and a graph clustering procedure which groups similar sequences into clusters. We demonstrate the viability of using the posteriorgram approach to handle many talkers by finding clusters of words in the TIMIT corpus.

show abstract

Keyword Spotting of Arbitrary Words Using Minimal Speech Resources

Cited by 42 publications

References 5 publications

Unsupervised spoken keyword spotting via segmental DTW on Gaussian posteriorgrams

Unsupervised spoken keyword spotting via segmental DTW on Gaussian posteriorgrams

Resource configurable spoken query detection using Deep Boltzmann Machines

Towards multi-speaker unsupervised speech pattern discovery

Contact Info

Product

Resources

About