The Experts below are selected from a list of 327 Experts worldwide ranked by ideXlab platform
M.-y. Hwang - One of the best experts on this subject based on the ideXlab platform.
-
from Sphinx ii to whisper making speech recognition usable
1996Co-Authors: Xuedong Huang, M.-y. Hwang, Fileno A Alleva, Alejandro Acero, Li Jiang, Milind MahajanAbstract:In this chapter, we first review Sphinx-II, a large-vocabulary speaker-independent continuous speech recognition system developed at Carnegie Mellon University, summarizing the techniques that helped Sphinx-II achieve state-of-the-art recognition performance. We then review Whisper, a system we developed here at Microsoft Corporation, focusing on recognition accuracy, efficiency and usability issues. These three issues are critical to the success of commercial speech applications. Whisper has significantly improved its performance in these three areas. It can be configured as a spoken language front-end (telephony or desktop) or dictation application.
-
an overview of the Sphinx ii speech recognition system
Human Language Technology, 1993Co-Authors: Xuedong Huang, M.-y. Hwang, Fileno A Alleva, Ronald RosenfeldAbstract:In the past year at Carnegie Mellon steady progress has been made in the area of acoustic and language modeling. The result has been a dramatic reduction in speech recognition errors in the Sphinx-II system. In this paper, we review Sphinx-II and summarize our recent efforts on improved speech recognition. Recently Sphinx-II achieved the lowest error rate in the November 1992 DARPA evaluations. For 5000-word, speaker-independent, continuous, speech recognition, the error rate was reduced to 5%.
-
HLT - An overview of the Sphinx-II speech recognition system
Proceedings of the workshop on Human Language Technology - HLT '93, 1993Co-Authors: Xuedong Huang, M.-y. Hwang, Fileno A Alleva, Ronald RosenfeldAbstract:In the past year at Carnegie Mellon steady progress has been made in the area of acoustic and language modeling. The result has been a dramatic reduction in speech recognition errors in the Sphinx-II system. In this paper, we review Sphinx-II and summarize our recent efforts on improved speech recognition. Recently Sphinx-II achieved the lowest error rate in the November 1992 DARPA evaluations. For 5000-word, speaker-independent, continuous, speech recognition, the error rate was reduced to 5%.
-
applying Sphinx ii to the darpa wall street journal csr task
Human Language Technology, 1992Co-Authors: F Alleva, M.-y. Hwang, Xuedong Huang, Roni Rosenfeld, R WeideAbstract:This paper reports recent efforts to apply the speaker-independent Sphinx-II system to the DARPA Wall Street Journal continuous speech recognition task. In Sphinx-II, we incorporated additional dynamic and speaker-normalized features, replaced discrete models with sex-dependent semi-continuous hidden Markov models, augmented within-word triphones with between-word triphones, and extended generalized triphone models to shared-distribution models. The configuration of Sphinx-II being used for this task includes sex-dependent, semi-continuous, shared-distribution hidden Markov models and left context dependent between-word triphones. In applying our technology to this task we addressed issues that were not previously of concern owing to the (relatively) small size of the Resource Management task.
-
Improved acoustic modeling with the Sphinx speech recognition system
[Proceedings] ICASSP 91: 1991 International Conference on Acoustics Speech and Signal Processing, 1991Co-Authors: X.d. Huang, M.-y. HwangAbstract:The authors report recent efforts to further improve the performance of the Sphinx system for speaker-independent continuous speech recognition. They adhere to the basic architecture of the Sphinx system and use the DARPA resource management task and training corpus. The improvements are evaluated on the 600 sentences that comprise the DARPA February and October 1989 test sets. Several techniques that substantially reduced Sphinx's error rate are presented. These techniques include dynamic features, semicontinuous hidden Markov models, speaker clustering, and the shared distribution modeling. The error rate of the baseline system was reduced by 45%.
R. Reddy - One of the best experts on this subject based on the ideXlab platform.
-
automatic speech recognition the development of the Sphinx system
2013Co-Authors: R. ReddyAbstract:1. Introduction.- 2. Hidden Markov Modeling of Speech.- 3. Task and Databases.- 4. The Baseline Sphinx System.- 5. Adding Knowledge.- 6. Finding a Good Unit of Speech.- 7. Learning and Adaptation.- 8. Summary of Results.- 9. Conclusion.- Appendix I. Evaluating Speech Recognizers.- I.1. Perplexity.- I.2. Computing Error Rate.- Appendix H. The Resource Management Task.- II.1. The Vocabulary and the Sphinx Pronunciation Dictionary.- II.2. The Grammar.- II.3. Training and Test Speakers.- Appendix III. Examples of Sphinx Recognition.- References.
-
An Overview of the Sphinx Speech Recognition System
IEEE Transactions on Acoustics Speech and Signal Processing, 1990Co-Authors: Kai-fu Lee, Hsiao-wuen Hon, R. ReddyAbstract:A description is given of Sphinx, a system that demonstrates the\nfeasibility of accurate, large-vocabulary, speaker-independent,\ncontinuous speech recognition. Sphinx is based on discrete hidden Markov\nmodels (HMMs) with LPC- (linear-predictive-coding) derived parameters.\nTo provide speaker independence, knowledge was added to these HMMs in\nseveral ways: multiple codebooks of fixed-width parameters, and an\nenhanced recognizer with carefully designed models and word-duration\nmodeling. To deal with coarticulation in continuous speech, yet still\nadequately represent a large vocabulary, two new subword speech units\nare introduced: function-word-dependent phone models and generalized\ntriphone models. With grammars of perplexity 997, 60, and 20, Sphinx\nattained word accuracies of 71, 94, and 96%, respectively, on a 997-word\ntask
Laura Manuelidis - One of the best experts on this subject based on the ideXlab platform.
-
prokaryotic Sphinx 1 8 rep protein is tissue specific and expressed in human germline cells
Journal of Cellular Biochemistry, 2019Co-Authors: Laura ManuelidisAbstract:: Small circular DNAs of 1.8 and 2.4 kb were initially discovered in highly infectious Creutzfeldt-Jakob Disease (CJD) and scrapie particles from mammalian brain and cultured cells. Surprisingly, these protected cytoplasmic "Sphinx" DNAs contained replication (REP) initiation sequences resembling those of Acinetobacter phage viruses. An antibody was generated against a REP peptide encoded by the Sphinx 1.8 open reading frame (ORF) that was not present in mammals. It bound to a 41kd "spx1" protein on Western blots. Cytologically, spx1 concentrated in spinal cord synapses and pancreatic islet, but not exocrine cells. We hypothesized that circular Sphinx DNAs are ancient symbiotic elements that can participate in functional differentiation and neurodegeneration. Cell and tissue-specific patterns of spx1 expression shown below implicate somatic cell-to-cell communication and differentiation functions that would favor conservation of Sphinx 1.8 in evolution. Remarkably, primary human oocytes and spermatogonia, but not mature sperm, displayed intense cytoplasmic spx1 signals that underscore the maternal inheritance of Sphinx 1.8. These findings should encourage investigations of unexplored networks of incorporated environmental infectious agents that can be key actors in progressive neurodegeneration, immunity, and cancer.
-
the prokaryotic Sphinx 1 8 rep protein is tissue specific and expressed in human germline cells
bioRxiv, 2018Co-Authors: Laura ManuelidisAbstract:Small circular DNAs of 1.8 and 2.4kb were initially discovered in highly infectious CJD and scrapie particles from mammalian brain and cultured cells. Surprisingly, these protected cytoplasmic "Sphinx" DNAs contained replication (REP) initiation sequences resembling those of Acinetobacter phage viruses. An antibody was generated against a REP peptide encoded by the Sphinx 1.8 ORF that was not present in mammals. It bound to a 41kd "spx1" protein on Western blots. Cytologically, spx1 concentrated in spinal cord synapses and pancreatic islet, but not exocrine cells. We hypothesized that circular Sphinx DNAs are ancient symbiotic elements that can participate in functional differentiation and neurodegeneration. Cell and tissue specific patterns of spx1 expression shown below implicate somatic cell-to-cell communication and differentiation functions that would favor conservation of Sphinx 1.8 in evolution. Remarkably, primary human oocytes and spermatogonia, but not mature sperm, displayed intense cytoplasmic spx1 signals that underscore the maternal inheritance of Sphinx 1.8. These findings should encourage investigations of unexplored networks of incorporated environmental infectious agents that can be key actors in progressive neurodegeneration, immunity and cancer.
-
Nuclease resistant circular DNAs copurify with infectivity in scrapie and CJD
Journal of NeuroVirology, 2011Co-Authors: Laura ManuelidisAbstract:In transmissible encephalopathies (TSEs), it is commonly believed that the host prion protein transforms itself into an infectious form that encodes the many distinct TSE agent strains without any nucleic acid. Using a Ф29 polymerase and chromatography strategy, highly infectious culture and brain preparations of three different geographic TSE agents all contained novel circular DNAs. Two circular “Sphinx” sequences, of 1.8 and 2.4 kb, copurified with infectious particles in sucrose gradients and, as many protected viruses, resisted nuclease digestion. Each contained a replicase ORF related to microviridae that infect commensal Acinetobacter. Infectious gradient fractions also contained nuclease-resistant 16 kb mitochondrial DNAs and analysis of >4,000 nt demonstrated a 100% identity with their species-specific sequences. This confirmed the fidelity of the newly identified sequences detailed here. Conserved replicase regions within the two Sphinx DNAs were ultimately detected by PCR in cytoplasmic preparations from normal cells and brain but were 2,500-fold less than in parallel-infected samples. No trace of the two Sphinx replicases was found in enzymes, detergents, or other preparative materials using exhaustive PCR cycles. The Sphinx sequences uncovered here could have a role in TSE infections despite their apparently symbiotic, low-level persistence in normal cells and tissues. These, as well as other cryptic circular DNAs, may cause or contribute to neurodegeneration and infection-associated tumor transformation. The current results also raise the intriguing possibility that mammals may incorporate more of the prokaryotic world in their cytoplasm than previously recognized.
Vijay Mann - One of the best experts on this subject based on the ideXlab platform.
-
NDSS - Sphinx: Detecting Security Attacks in Software-Defined Networks.
Proceedings 2015 Network and Distributed System Security Symposium, 2020Co-Authors: Mohan Dhawan, Rishabh Poddar, Kshiteej Mahajan, Vijay MannAbstract:Software-defined networks (SDNs) allow greater control over network entities by centralizing the control plane, but place great burden on the administrator to manually ensure security and correct functioning of the entire network. We list several attacks on SDN controllers that violate network topology and data plane forwarding, and can be mounted by compromised network entities, such as end hosts and soft switches. We further demonstrate their feasibility on four popular SDN controllers. We propose Sphinx to detect both known and potentially unknown attacks on network topology and data plane forwarding originating within an SDN. Sphinx leverages the novel abstraction of flow graphs, which closely approximate the actual network operations, to enable incremental validation of all network updates and constraints. Sphinx dynamically learns new network behavior and raises alerts when it detects suspicious changes to existing network control plane behavior. Our evaluation shows that Sphinx is capable of detecting attacks in SDNs in realtime with low performance overheads, and requires no changes to the controllers for deployment.
-
Sphinx detecting security attacks in software defined networks
Network and Distributed System Security Symposium, 2015Co-Authors: Mohan Dhawan, Rishabh Poddar, Kshiteej Mahajan, Vijay MannAbstract:Software-defined networks (SDNs) allow greater control over network entities by centralizing the control plane, but place great burden on the administrator to manually ensure security and correct functioning of the entire network. We list several attacks on SDN controllers that violate network topology and data plane forwarding, and can be mounted by compromised network entities, such as end hosts and soft switches. We further demonstrate their feasibility on four popular SDN controllers. We propose Sphinx to detect both known and potentially unknown attacks on network topology and data plane forwarding originating within an SDN. Sphinx leverages the novel abstraction of flow graphs, which closely approximate the actual network operations, to enable incremental validation of all network updates and constraints. Sphinx dynamically learns new network behavior and raises alerts when it detects suspicious changes to existing network control plane behavior. Our evaluation shows that Sphinx is capable of detecting attacks in SDNs in realtime with low performance overheads, and requires no changes to the controllers for deployment.
Mohan Dhawan - One of the best experts on this subject based on the ideXlab platform.
-
NDSS - Sphinx: Detecting Security Attacks in Software-Defined Networks.
Proceedings 2015 Network and Distributed System Security Symposium, 2020Co-Authors: Mohan Dhawan, Rishabh Poddar, Kshiteej Mahajan, Vijay MannAbstract:Software-defined networks (SDNs) allow greater control over network entities by centralizing the control plane, but place great burden on the administrator to manually ensure security and correct functioning of the entire network. We list several attacks on SDN controllers that violate network topology and data plane forwarding, and can be mounted by compromised network entities, such as end hosts and soft switches. We further demonstrate their feasibility on four popular SDN controllers. We propose Sphinx to detect both known and potentially unknown attacks on network topology and data plane forwarding originating within an SDN. Sphinx leverages the novel abstraction of flow graphs, which closely approximate the actual network operations, to enable incremental validation of all network updates and constraints. Sphinx dynamically learns new network behavior and raises alerts when it detects suspicious changes to existing network control plane behavior. Our evaluation shows that Sphinx is capable of detecting attacks in SDNs in realtime with low performance overheads, and requires no changes to the controllers for deployment.
-
Sphinx detecting security attacks in software defined networks
Network and Distributed System Security Symposium, 2015Co-Authors: Mohan Dhawan, Rishabh Poddar, Kshiteej Mahajan, Vijay MannAbstract:Software-defined networks (SDNs) allow greater control over network entities by centralizing the control plane, but place great burden on the administrator to manually ensure security and correct functioning of the entire network. We list several attacks on SDN controllers that violate network topology and data plane forwarding, and can be mounted by compromised network entities, such as end hosts and soft switches. We further demonstrate their feasibility on four popular SDN controllers. We propose Sphinx to detect both known and potentially unknown attacks on network topology and data plane forwarding originating within an SDN. Sphinx leverages the novel abstraction of flow graphs, which closely approximate the actual network operations, to enable incremental validation of all network updates and constraints. Sphinx dynamically learns new network behavior and raises alerts when it detects suspicious changes to existing network control plane behavior. Our evaluation shows that Sphinx is capable of detecting attacks in SDNs in realtime with low performance overheads, and requires no changes to the controllers for deployment.
