BYBLOS: The BBN continuous speech recognition system

Chow, Yen‐Lu; Dunham, M.; Kimball, Owen; Krasner, M.; Kubala, G.; Makhoul, John; Price, Patti; Roucos, S.; Schwartz, Richard

doi:10.1109/icassp.1987.1169748

Cited by 112 publications

(20 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The basic BBN Byblos system is essentially the same as originally described in [2]. These experiments used contextdependent but not cross-word triphone models.…”

Section: Resultsmentioning

confidence: 99%

Integration of diverse recognition methodologies through reevaluation of N-best sentence hypotheses

Ostendorf¹,

Kannan²,

Austin³

et al. 1991

Proceedings of the Workshop on Speech and Natural Language - HLT '91

Self Cite

View full text Add to dashboard Cite

This paper describes a general formalism for integrating two or more speech recognition technologies, which could be developed at different research sites using different recognition strategies. In this formalism, one system uses the N-best search strategy to generate a list of candidate sentences; the list is rescorred by other systems; and the different scores axe combined to optimize performance. Specifically, we report on combining the BU system based on stochastic segment models and the BBN system based on hidden Markov models. In addition to facilitating integration of different systems, the N-best approach results in a large reduction in computation for word recognition using the stochastic segment model

show abstract

“…The basic BBN Byblos system is essentially the same as originally described in [2]. These experiments used contextdependent but not cross-word triphone models.…”

Section: Resultsmentioning

confidence: 99%

Integration of diverse recognition methodologies through reevaluation of N-best sentence hypotheses

Ostendorf¹,

Kannan²,

Austin³

et al. 1991

Proceedings of the Workshop on Speech and Natural Language - HLT '91

Self Cite

View full text Add to dashboard Cite

show abstract

“…Sage is BBN's newly developed STT platform that integrates technologies from multiple sources, each of which has a particular strength. In Sage, we combine proprietary sources, such as BBN's Byblos [24], with open source toolkits, such as Kaldi [25], CNTK [26] and Tensorflow [27]. For example, DNN can be trained using Byblos, Kaldi nnet1 [28] or nnet2, CNN using Kaldi or Caffé [29], and LSTM using Kaldi or CNTK.…”

Section: Methodsmentioning

confidence: 99%

Two-Stage Data Augmentation for Low-Resourced Speech Recognition

Hartmann¹,

Ng²,

Hsiao³

et al. 2016

Interspeech 2016

View full text Add to dashboard Cite

Low resourced languages suffer from limited training data and resources. Data augmentation is a common approach to increasing the amount of training data. Additional data is synthesized by manipulating the original data with a variety of methods. Unlike most previous work that focuses on a single technique, we combine multiple, complementary augmentation approaches. The first stage adds noise and perturbs the speed of additional copies of the original audio. The data is further augmented in a second stage, where a novel fMLLR-based augmentation is applied to bottleneck features to further improve performance. A reduction in word error rate is demonstrated on four languages from the IARPA Babel program. We present an analysis exploring why these techniques are beneficial. Index Terms: speech recognition, deep neural networks, data augmentation Speed PerturbationKo et al. [18] showed success by manipulating the speed of the data. They demonstrated a performance improvement over the more common vocal tract length perturbation (VTLP) technique [8]. Using the Sox utility [19], the original data is perturbed by a warping factor that effects both the frequencies and the duration of the speech. The speed change is accomplished by resampling the waveform, which not only changes the duration, but also scales the pitch, vocal tract length, and all spectral frequencies by the same factor. Our setup uses a randomly selected warping factor between 0.9 and 1.1 (this was also the

show abstract

“…For extracting the highlights, we developed a HMM model which takes the initial probabilities and the transition probability matrix of the states as input [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23]. The steps involved in the highlight extraction process are described below.…”

Section: Extraction Of the Highlightsmentioning

confidence: 99%

Automatic extraction of highlights from a cricket video using MPEG-7 descriptors

Namuduri

2009

2009 First International Communication Systems and Networks and Workshops

View full text Add to dashboard Cite

Sports videos appeal to a vast majority of people. Hence, video streaming of sports videos presents a profitable business opportunity. As a result, there is an astronomical increase in the amount of sports content available on the internet. Businesses and users face a new problem of perusing such large amounts of multimedia content quickly for the required information. One technique that can enable quick overview of multimedia content is video summarization.In this paper, an approach to summarize a cricket game video by using highlights extraction process is presented. The video is first segmented into shots. Key frames are extracted from each shot and low level features are extracted from the key frames. Feature are used to extract views or states. The states and their transitions in the cricket game are represented by a Hidden Markov Model, based on which the game highlights are extracted. The performance of the proposed summarization scheme is evaluated for accuracy.

show abstract

BYBLOS: The BBN continuous speech recognition system

Cited by 112 publications

References 5 publications

Integration of diverse recognition methodologies through reevaluation of N-best sentence hypotheses

Integration of diverse recognition methodologies through reevaluation of N-best sentence hypotheses

Two-Stage Data Augmentation for Low-Resourced Speech Recognition

Automatic extraction of highlights from a cricket video using MPEG-7 descriptors

Contact Info

Product

Resources

About