H-Index & Metrics Top Publications

H-Index & Metrics

Discipline name H-index Citations Publications World Ranking National Ranking
Computer Science H-index 61 Citations 26,022 158 World Ranking 1473 National Ranking 139

Overview

What is he best known for?

The fields of study he is best known for:

  • Artificial intelligence
  • Statistics
  • Speech recognition

Daniel Povey mostly deals with Speech recognition, Artificial intelligence, Pattern recognition, Artificial neural network and Hidden Markov model. His Speech recognition study incorporates themes from Reduction and Robustness. Many of his studies on Artificial intelligence involve topics that are commonly interrelated, such as Machine learning.

His Pattern recognition research focuses on Gaussian process and how it connects with Feature vector. His biological study spans a wide range of topics, including Speaker verification, Speaker recognition, Lattice, Topology and Deep learning. His Mixture model research is multidisciplinary, incorporating perspectives in Subspace topology and Subspace Gaussian Mixture Model.

His most cited work include:

  • The Kaldi Speech Recognition Toolkit (3765 citations)
  • Librispeech: An ASR corpus based on public domain audio books (1658 citations)
  • X-Vectors: Robust DNN Embeddings for Speaker Recognition (869 citations)

What are the main themes of his work throughout his whole career to date?

Daniel Povey focuses on Speech recognition, Artificial intelligence, Artificial neural network, Hidden Markov model and Word error rate. His Speech recognition research is multidisciplinary, relying on both Discriminative model and Mutual information. His study in Artificial intelligence is interdisciplinary in nature, drawing from both Natural language processing, Machine learning and Pattern recognition.

His work on Time delay neural network as part of general Artificial neural network study is frequently connected to Adaptation, therefore bridging the gap between diverse disciplines of science and establishing a new relationship between them. His Word error rate research is multidisciplinary, incorporating elements of Transcription, Vocal tract, NIST and Test set. His Mixture model study combines topics from a wide range of disciplines, such as Subspace topology and Subspace Gaussian Mixture Model.

He most often published in these fields:

  • Speech recognition (71.65%)
  • Artificial intelligence (41.24%)
  • Artificial neural network (23.20%)

What were the highlights of his more recent work (between 2018-2021)?

  • Speech recognition (71.65%)
  • Artificial neural network (23.20%)
  • Language model (13.40%)

In recent papers he was focusing on the following fields of study:

His primary areas of study are Speech recognition, Artificial neural network, Language model, Speaker diarisation and Decoding methods. He works in the field of Speech recognition, namely Speaker recognition. Artificial intelligence covers Daniel Povey research in Artificial neural network.

The Language model study combines topics in areas such as Algorithm, Recurrent neural network and Vocabulary. He interconnects Speech enhancement and Conversational speech in the investigation of issues within Speaker diarisation. His work carried out in the field of Decoding methods brings together such families of science as Speedup and Parallel computing.

Between 2018 and 2021, his most popular works were:

  • Speaker Recognition for Multi-speaker Conversations Using X-vectors (94 citations)
  • CHiME-6 Challenge:Tackling Multispeaker Speech Recognition for Unsegmented Recordings (49 citations)
  • State-of-the-Art Speaker Recognition for Telephone and Video Speech: The JHU-MIT Submission for NIST SRE18. (42 citations)

In his most recent research, the most cited papers focused on:

  • Artificial intelligence
  • Statistics
  • Speech recognition

The scientist’s investigation covers issues in Speech recognition, Speaker recognition, Speaker diarisation, Artificial neural network and Time delay neural network. Daniel Povey integrates many fields, such as Speech recognition and Set, in his works. His study in the field of Speaker verification also crosses realms of Extractor.

The study incorporates disciplines such as Speech enhancement and Conversational speech in addition to Speaker diarisation. The various areas that Daniel Povey examines in his Artificial neural network study include Pipeline and Training set. His Time delay neural network research incorporates themes from Vocabulary, Arabic, Convolutional neural network, Hidden Markov model and Machine translation.

This overview was generated by a machine learning system which analysed the scientist’s body of work. If you have any feedback, you can contact us here.

Top Publications

The Kaldi Speech Recognition Toolkit

Daniel Povey;Arnab Ghoshal;Gilles Boulianne;Lukas Burget.
ieee automatic speech recognition and understanding workshop (2011)

5086 Citations

Librispeech: An ASR corpus based on public domain audio books

Vassil Panayotov;Guoguo Chen;Daniel Povey;Sanjeev Khudanpur.
international conference on acoustics, speech, and signal processing (2015)

1868 Citations

X-Vectors: Robust DNN Embeddings for Speaker Recognition

David Snyder;Daniel Garcia-Romero;Gregory Sell;Daniel Povey.
international conference on acoustics, speech, and signal processing (2018)

1099 Citations

The HTK book version 3.4

SJ Young;G Evermann;Mjf Gales;D Kershaw.
(2006)

977 Citations

Minimum Phone Error and I-smoothing for improved discriminative training

D. Povey;P.C. Woodland.
international conference on acoustics, speech, and signal processing (2002)

895 Citations

Sequence-discriminative training of deep neural networks

Karel Veselý;Arnab Ghoshal;Lukás Burget;Daniel Povey.
conference of the international speech communication association (2013)

782 Citations

A time delay neural network architecture for efficient modeling of long temporal contexts.

Vijayaditya Peddinti;Daniel Povey;Sanjeev Khudanpur.
conference of the international speech communication association (2015)

766 Citations

Purely Sequence-Trained Neural Networks for ASR Based on Lattice-Free MMI.

Daniel Povey;Vijayaditya Peddinti;Daniel Galvez;Pegah Ghahremani.
conference of the international speech communication association (2016)

631 Citations

Audio augmentation for speech recognition.

Tom Ko;Vijayaditya Peddinti;Daniel Povey;Sanjeev Khudanpur.
conference of the international speech communication association (2015)

618 Citations

Deep Neural Network Embeddings for Text-Independent Speaker Verification.

David Snyder;Daniel Garcia-Romero;Daniel Povey;Sanjeev Khudanpur.
conference of the international speech communication association (2017)

547 Citations

Profile was last updated on December 6th, 2021.
Research.com Ranking is based on data retrieved from the Microsoft Academic Graph (MAG).
The ranking h-index is inferred from publications deemed to belong to the considered discipline.

If you think any of the details on this page are incorrect, let us know.

Contact us

Top Scientists Citing Daniel Povey

Shinji Watanabe

Shinji Watanabe

Carnegie Mellon University

Publications: 134

Hermann Ney

Hermann Ney

RWTH Aachen University

Publications: 106

Mark J. F. Gales

Mark J. F. Gales

University of Cambridge

Publications: 102

Haizhou Li

Haizhou Li

Chinese University of Hong Kong, Shenzhen

Publications: 101

Dong Yu

Dong Yu

Tencent (China)

Publications: 99

Steve Renals

Steve Renals

University of Edinburgh

Publications: 87

Ralf Schlüter

Ralf Schlüter

RWTH Aachen University

Publications: 80

James Glass

James Glass

MIT

Publications: 79

Chin-Hui Lee

Chin-Hui Lee

Georgia Institute of Technology

Publications: 73

Lukas Burget

Lukas Burget

Brno University of Technology

Publications: 71

Bhuvana Ramabhadran

Bhuvana Ramabhadran

Google (United States)

Publications: 71

Florian Metze

Florian Metze

Carnegie Mellon University

Publications: 63

Philip C. Woodland

Philip C. Woodland

University of Cambridge

Publications: 62

Shrikanth S. Narayanan

Shrikanth S. Narayanan

University of Southern California

Publications: 57

Jinyu Li

Jinyu Li

Microsoft (United States)

Publications: 57

Something went wrong. Please try again later.