Skip to main content

nada nada

Followers

0

Following

1

Public Views

University of Edinburgh

Bar-Ilan University

Christopher D Manning

Stanford University

Interests

Uploads

Papers by nada nada

Estimation of probabilities in the language model of the IBM speech recognition system

IEEE Transactions on Acoustics, Speech, and Signal Processing, 1984

The language model probabilities are estimated by an empirical Bayes approach in which a prior di... more The language model probabilities are estimated by an empirical Bayes approach in which a prior distribution for the unknown probabilities is itself estimated through a novel choice of data. The predictive power of the model thus fitted is compared by means of its experimental perplexity [l] to the model as fitted by the Jelinek-Mercer deleted estimator and as fitted by the Turing-Good formulas for probabilities of unseen or rarely seen events.

Continuous speech recognition with automatically selected acoustic prototypes obtained by either bootstrapping or clustering

ICASSP '81. IEEE International Conference on Acoustics, Speech, and Signal Processing

Automatic selection of acoustic prototypes is an important step towards making speech recognition... more Automatic selection of acoustic prototypes is an important step towards making speech recognition systems automatically adaptable to new speakers. Two methods for automatically obtaining a set of acoustic prototypes for use by a centisecond labeling acoustic processor are described. One method is based on bootstrapping, the other on clustering. Recognition results using these automatically obtained prototypes on the 1000-word vocabulary

Políticas de Controle da Desordem Urbana: A Experiência das Unidades de Ordem Pública na Cidade do Rio de Janeiro

Revista de Cultura e Extensão USP, 2015

Este artigo descreve e analisa o processo de desenho e execução das Unidades de Ordem Pública (UO... more

Large vocabulary natural language continuous speech recognition

International Conference on Acoustics, Speech, and Signal Processing

The present paper describes our current research on automatic speech recognition of continuously ... more The present paper describes our current research on automatic speech recognition of continuously read sentences from a naturally-occurring corpus: office correspondence. The recognition system combines features from our current isolated-word recognition system and from our previously developed continuous speech recognition systems. It consists of an acoustic processor, an acoustic channel model, a language model, and a linguistic decoder. Some new features in the recognizer relative to our isolated-word speech recognition system include the use of a fast match to rapidly prune to a manageable number the candidates considered by the detailed match, multiple pronunciations of all function words, and modelling of interphone coarticulatory behavior_ To date, we have recorded training and test data from a set of 10 male talkers. The test data consist of 50 sentences drawn from spontaneously generated memos covered by a 5000 word vocabulary. The perplexity of the test sentences was found to be 93; none of the sentences were part of the data used to generate the language model. Preliminary (speaker-dependent) recognition results on these talkers yielded an average word error rate of 11-0%.

Speech recognition using noise-adaptive prototypes

ICASSP-88., International Conference on Acoustics, Speech, and Signal Processing

A probabilistic mixture model is described for a frame (the short-term spectrum) of each to be us... more A probabilistic mixture model is described for a frame (the short-term spectrum) of each to be used in speech recognition. Each component of the mixture is regarded as a prototype for the labeling phase of a hidden Markov model based speech recognition system. Since the ambient noise during recognition can differ from the ambient noise present in the training data, the model is designed for convenient updating in changing noise. Based on the observation that the energy in a frequency band is at any fixed time dominated either by signal energy or by noise energy, the authors model the energy as the larger of the separate energies of signal and noise in the band. Statistical algorithms are given for training this as a hidden variables model. The hidden variables are the prototype identities and the separate signal and noise components. A series of speech recognition experiments that successfully utilize this model is also discussed.<<ETX>>

Accelerating Kirchhoff Migration by CPU and GPU Cooperation

2009 21st International Symposium on Computer Architecture and High Performance Computing, 2009

We discuss the performance of Petrobras production Kirchhoff prestack seismic migration on a clus... more We discuss the performance of Petrobras production Kirchhoff prestack seismic migration on a cluster of 64 GPUs and 256 CPU cores. Porting and optimization of the application hot spot (98.2% of a single CPU core execution time) to a single GPU reduces total execution time by a factor of 36 on a control run. We then argue against the usual practice of porting the next hot spot (1.5% of single CPU core execution time) to the GPU. Instead, we show that cooperation of CPU and GPU reduces total execution time by a factor of 59 on the same control run. Remaining GPU idle cycles are eliminated by overloading the GPU with multiple requests originated from distinct CPU cores. However, increasing the number of CPU cores in the computation reduces the gain due to the combination of enhanced parallelism in the runs without GPUs and GPU saturation on runs with GPUs. We proceed by obtaining close to perfect speed-up on the full cluster over homogeneous load obtained by replicating control run data. To cope with the heterogeneous load of real world data we show a dynamic load balancing scheme that reduces total execution time by a factor of 20 on runs that use all GPUs and half of the cluster CPU cores with respect to runs that use all CPU cores but no GPU. Prestack Kirchhoff time migration has been used by the seismic industry for a long time. In a previous work

Seismic imaging on novel computer architectures

11th International Congress of the Brazilian Geophysical Society & EXPOGEF 2009, Salvador, Bahia, Brazil, 24-28 August 2009, 2009

This paper describes performance results obtained by porting production seismic imaging software ... more

Accelerating Time and Depth Seismic Migration by CPU and GPU Cooperation

International Journal of Parallel Programming, 2011

In this paper, we discuss the porting strategy and the achieved performance of Petrobras producti... more In this paper, we discuss the porting strategy and the achieved performance of Petrobras production code that implements Kirchhoff prestack time and depth seismic migration algorithms on a cluster of 64 GPUs and 256 CPU cores. The time migration of a real-world survey was 20 times faster in the full cluster with GPUs than in the full cluster without GPUs. The introduction of GPUs accelerated the depth migration of another real-world survey by a factor of 5. In both cases, a large fraction of the achieved gains were due to the porting strategy that privileges CPU-GPU A preliminary version of this work was published at the 21st International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2009.

Liderança local, democracia e políticas públicas no Brasil

Opinião Pública, 2003

Este artigo aborda a questão democrática enfocando a dinâmica da política local no Brasil. À luz ... more Este artigo aborda a questão democrática enfocando a dinâmica da política local no Brasil. À luz das evidências políticas que embasaram o importante papel do município como alicerce da democracia, estabelecida pelo modelo federativo resultante da Constituição de 1988, como a dinâmica de transferência de recursos e o desenho das políticas sociais, o trabalho analisa de forma geral o mapa de valores e opiniões de elites e lideranças locais com relação a atribuições e competências de diferentes níveis de governo. Os dados analisados provêm da pesquisa Democracia e governo local, realizada com 450 lideranças de municípios brasileiros.

Definindo a arena política local: sistemas partidários municipais na federação brasileira

Dados, 2008

E ste artigo trata da conexão entre a arena política local e as arenas estaduais e nacional consi... more E ste artigo trata da conexão entre a arena política local e as arenas estaduais e nacional considerando o prisma do sistema partidário. Quisemos verificar em que medida e de que forma a dinâmica político-eleitoral nos municípios é tributária do que se passa nas outras esferas da Federação. A experiência democrática brasileira não pode ser adequadamente estudada sem se levar em conta que ela ocorre nos marcos de um sistema federativo. A Federação brasileira tem mais de cem anos e constitui uma das instituições mais relevantes para o entendimento da estruturação do sistema político brasileiro e dos processos que o movimentam. Entretanto, só recentemente se tornou objeto da atenção sistemática dos cientistas políticos. Estudos recentes têm contribuído para re-403 * Este texto é um dos produtos finais da pesquisa Democracia, Política e Governo Local, projeto temático da Fundação de Amparo à Pesquisa do Estado de São Paulo-FAPESP. Em versão muito preliminar, foi discutido no seminário da área

Pesquisas De Vitimização e Gestão Da Segurança Pública

Esse artigo analisa como as pesquisas de vitimização podem contribuir para o aumento da eficiênci... more Esse artigo analisa como as pesquisas de vitimização podem contribuir para o aumento da eficiência e da eficácia das intervenções públicas, com vistas à redução do crime. Em particular, avalia de que forma esses levantamentos podem complementar os demais sistemas de informações sobre a segurança pública e discute os aspectos metodológicos, que têm efeito sobre as estimativas do número de vítimas e das taxas de recurso à polícia Palavras-chave: Pesquisas de vitimização. Aspectos metodológicos de pesquisas de vitimização. Segurança pública. Controle do crime.

Democratic consolidation and civil rights: Brazil in comparative perspective

Chapter published at: &quot;Brazil since 1985: Politics, Economy and Society&quot;, by Ma... more

Estimation of probabilities in the language model of the IBM speech recognition system

IEEE Transactions on Acoustics, Speech, and Signal Processing, 1984

The language model probabilities are estimated by an empirical Bayes approach in which a prior di... more The language model probabilities are estimated by an empirical Bayes approach in which a prior distribution for the unknown probabilities is itself estimated through a novel choice of data. The predictive power of the model thus fitted is compared by means of its experimental perplexity [l] to the model as fitted by the Jelinek-Mercer deleted estimator and as fitted by the Turing-Good formulas for probabilities of unseen or rarely seen events.

Continuous speech recognition with automatically selected acoustic prototypes obtained by either bootstrapping or clustering

ICASSP '81. IEEE International Conference on Acoustics, Speech, and Signal Processing

Automatic selection of acoustic prototypes is an important step towards making speech recognition... more Automatic selection of acoustic prototypes is an important step towards making speech recognition systems automatically adaptable to new speakers. Two methods for automatically obtaining a set of acoustic prototypes for use by a centisecond labeling acoustic processor are described. One method is based on bootstrapping, the other on clustering. Recognition results using these automatically obtained prototypes on the 1000-word vocabulary

Políticas de Controle da Desordem Urbana: A Experiência das Unidades de Ordem Pública na Cidade do Rio de Janeiro

Revista de Cultura e Extensão USP, 2015

Este artigo descreve e analisa o processo de desenho e execução das Unidades de Ordem Pública (UO... more

Large vocabulary natural language continuous speech recognition

International Conference on Acoustics, Speech, and Signal Processing

The present paper describes our current research on automatic speech recognition of continuously ... more The present paper describes our current research on automatic speech recognition of continuously read sentences from a naturally-occurring corpus: office correspondence. The recognition system combines features from our current isolated-word recognition system and from our previously developed continuous speech recognition systems. It consists of an acoustic processor, an acoustic channel model, a language model, and a linguistic decoder. Some new features in the recognizer relative to our isolated-word speech recognition system include the use of a fast match to rapidly prune to a manageable number the candidates considered by the detailed match, multiple pronunciations of all function words, and modelling of interphone coarticulatory behavior_ To date, we have recorded training and test data from a set of 10 male talkers. The test data consist of 50 sentences drawn from spontaneously generated memos covered by a 5000 word vocabulary. The perplexity of the test sentences was found to be 93; none of the sentences were part of the data used to generate the language model. Preliminary (speaker-dependent) recognition results on these talkers yielded an average word error rate of 11-0%.

Speech recognition using noise-adaptive prototypes

ICASSP-88., International Conference on Acoustics, Speech, and Signal Processing

A probabilistic mixture model is described for a frame (the short-term spectrum) of each to be us... more A probabilistic mixture model is described for a frame (the short-term spectrum) of each to be used in speech recognition. Each component of the mixture is regarded as a prototype for the labeling phase of a hidden Markov model based speech recognition system. Since the ambient noise during recognition can differ from the ambient noise present in the training data, the model is designed for convenient updating in changing noise. Based on the observation that the energy in a frequency band is at any fixed time dominated either by signal energy or by noise energy, the authors model the energy as the larger of the separate energies of signal and noise in the band. Statistical algorithms are given for training this as a hidden variables model. The hidden variables are the prototype identities and the separate signal and noise components. A series of speech recognition experiments that successfully utilize this model is also discussed.<<ETX>>

Accelerating Kirchhoff Migration by CPU and GPU Cooperation

2009 21st International Symposium on Computer Architecture and High Performance Computing, 2009

We discuss the performance of Petrobras production Kirchhoff prestack seismic migration on a clus... more We discuss the performance of Petrobras production Kirchhoff prestack seismic migration on a cluster of 64 GPUs and 256 CPU cores. Porting and optimization of the application hot spot (98.2% of a single CPU core execution time) to a single GPU reduces total execution time by a factor of 36 on a control run. We then argue against the usual practice of porting the next hot spot (1.5% of single CPU core execution time) to the GPU. Instead, we show that cooperation of CPU and GPU reduces total execution time by a factor of 59 on the same control run. Remaining GPU idle cycles are eliminated by overloading the GPU with multiple requests originated from distinct CPU cores. However, increasing the number of CPU cores in the computation reduces the gain due to the combination of enhanced parallelism in the runs without GPUs and GPU saturation on runs with GPUs. We proceed by obtaining close to perfect speed-up on the full cluster over homogeneous load obtained by replicating control run data. To cope with the heterogeneous load of real world data we show a dynamic load balancing scheme that reduces total execution time by a factor of 20 on runs that use all GPUs and half of the cluster CPU cores with respect to runs that use all CPU cores but no GPU. Prestack Kirchhoff time migration has been used by the seismic industry for a long time. In a previous work

Seismic imaging on novel computer architectures

11th International Congress of the Brazilian Geophysical Society & EXPOGEF 2009, Salvador, Bahia, Brazil, 24-28 August 2009, 2009

This paper describes performance results obtained by porting production seismic imaging software ... more

Accelerating Time and Depth Seismic Migration by CPU and GPU Cooperation

International Journal of Parallel Programming, 2011

In this paper, we discuss the porting strategy and the achieved performance of Petrobras producti... more In this paper, we discuss the porting strategy and the achieved performance of Petrobras production code that implements Kirchhoff prestack time and depth seismic migration algorithms on a cluster of 64 GPUs and 256 CPU cores. The time migration of a real-world survey was 20 times faster in the full cluster with GPUs than in the full cluster without GPUs. The introduction of GPUs accelerated the depth migration of another real-world survey by a factor of 5. In both cases, a large fraction of the achieved gains were due to the porting strategy that privileges CPU-GPU A preliminary version of this work was published at the 21st International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2009.

Liderança local, democracia e políticas públicas no Brasil

Opinião Pública, 2003

Este artigo aborda a questão democrática enfocando a dinâmica da política local no Brasil. À luz ... more Este artigo aborda a questão democrática enfocando a dinâmica da política local no Brasil. À luz das evidências políticas que embasaram o importante papel do município como alicerce da democracia, estabelecida pelo modelo federativo resultante da Constituição de 1988, como a dinâmica de transferência de recursos e o desenho das políticas sociais, o trabalho analisa de forma geral o mapa de valores e opiniões de elites e lideranças locais com relação a atribuições e competências de diferentes níveis de governo. Os dados analisados provêm da pesquisa Democracia e governo local, realizada com 450 lideranças de municípios brasileiros.

Definindo a arena política local: sistemas partidários municipais na federação brasileira

Dados, 2008

E ste artigo trata da conexão entre a arena política local e as arenas estaduais e nacional consi... more E ste artigo trata da conexão entre a arena política local e as arenas estaduais e nacional considerando o prisma do sistema partidário. Quisemos verificar em que medida e de que forma a dinâmica político-eleitoral nos municípios é tributária do que se passa nas outras esferas da Federação. A experiência democrática brasileira não pode ser adequadamente estudada sem se levar em conta que ela ocorre nos marcos de um sistema federativo. A Federação brasileira tem mais de cem anos e constitui uma das instituições mais relevantes para o entendimento da estruturação do sistema político brasileiro e dos processos que o movimentam. Entretanto, só recentemente se tornou objeto da atenção sistemática dos cientistas políticos. Estudos recentes têm contribuído para re-403 * Este texto é um dos produtos finais da pesquisa Democracia, Política e Governo Local, projeto temático da Fundação de Amparo à Pesquisa do Estado de São Paulo-FAPESP. Em versão muito preliminar, foi discutido no seminário da área

Pesquisas De Vitimização e Gestão Da Segurança Pública

Esse artigo analisa como as pesquisas de vitimização podem contribuir para o aumento da eficiênci... more Esse artigo analisa como as pesquisas de vitimização podem contribuir para o aumento da eficiência e da eficácia das intervenções públicas, com vistas à redução do crime. Em particular, avalia de que forma esses levantamentos podem complementar os demais sistemas de informações sobre a segurança pública e discute os aspectos metodológicos, que têm efeito sobre as estimativas do número de vítimas e das taxas de recurso à polícia Palavras-chave: Pesquisas de vitimização. Aspectos metodológicos de pesquisas de vitimização. Segurança pública. Controle do crime.

Democratic consolidation and civil rights: Brazil in comparative perspective

Chapter published at: &quot;Brazil since 1985: Politics, Economy and Society&quot;, by Ma... more