H-Index & Metrics Best Publications

H-Index & Metrics

Discipline name H-index Citations Publications World Ranking National Ranking
Computer Science D-index 57 Citations 17,937 299 World Ranking 1960 National Ranking 17

Overview

What is he best known for?

The fields of study he is best known for:

  • Artificial intelligence
  • Speech recognition
  • Statistics

Speech recognition, Hidden Markov model, Speech synthesis, Artificial intelligence and Pattern recognition are his primary areas of study. His studies deal with areas such as Quality and Audio signal as well as Speech recognition. Keiichi Tokuda combines subjects such as Probability distribution, Hidden semi-Markov model and White noise with his study of Hidden Markov model.

His study in Speech synthesis is interdisciplinary in nature, drawing from both Speaker recognition, Speaker diarisation, Parametric statistics, Active listening and Sentence. His work deals with themes such as Duration, State and Natural language processing, which intersect with Artificial intelligence. His Pattern recognition study combines topics from a wide range of disciplines, such as Estimation theory and Set.

His most cited work include:

  • Statistical Parametric Speech Synthesis (953 citations)
  • Speech parameter generation algorithms for HMM-based speech synthesis (810 citations)
  • Voice Conversion Based on Maximum-Likelihood Estimation of Spectral Parameter Trajectory (765 citations)

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

Keiichi Tokuda mainly investigates Speech recognition, Hidden Markov model, Artificial intelligence, Speech synthesis and Pattern recognition. Speech recognition is represented through his Speaker recognition, Speaker diarisation, Speech coding, Speech processing and Voice activity detection research. His research investigates the link between Hidden Markov model and topics such as Cluster analysis that cross with problems in Decision tree.

He frequently studies issues relating to Natural language processing and Artificial intelligence. Keiichi Tokuda studied Speech synthesis and Parametric statistics that intersect with Artificial neural network. His Pattern recognition research integrates issues from Hidden semi-Markov model, Feature, Estimation theory and Markov model.

He most often published in these fields:

  • Speech recognition (68.82%)
  • Hidden Markov model (52.15%)
  • Artificial intelligence (48.92%)

What were the highlights of his more recent work (between 2013-2021)?

  • Speech recognition (68.82%)
  • Speech synthesis (42.20%)
  • Artificial intelligence (48.92%)

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

His primary areas of investigation include Speech recognition, Speech synthesis, Artificial intelligence, Hidden Markov model and Artificial neural network. His Singing voice synthesis study, which is part of a larger body of work in Speech recognition, is frequently linked to Naturalness, bridging the gap between disciplines. His Speech synthesis study integrates concerns from other disciplines, such as Feature, Active listening, End-to-end principle, Human–computer interaction and Generative grammar.

His Artificial intelligence research includes elements of Natural language processing and Pattern recognition. His Hidden Markov model research includes themes of Data modeling, Probability distribution and Feature extraction, Computer vision. His research integrates issues of Parametric statistics, Speech processing and Waveguide in his study of Artificial neural network.

Between 2013 and 2021, his most popular works were:

  • Singing voice synthesis based on deep neural networks (45 citations)
  • The effect of neural networks in statistical parametric speech synthesis (35 citations)
  • Singing Voice Synthesis Based on Generative Adversarial Networks (23 citations)

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

  • Artificial intelligence
  • Statistics
  • Speech recognition

His primary areas of study are Speech recognition, Artificial neural network, Hidden Markov model, Speech synthesis and Singing voice synthesis. His Speech recognition study combines topics in areas such as Singing and Feature extraction, Artificial intelligence. The concepts of his Artificial intelligence study are interwoven with issues in Natural language processing and Pattern recognition.

The Artificial neural network study combines topics in areas such as Waveform, Parametric statistics and Speech processing. Keiichi Tokuda brings together Hidden Markov model and Pronunciation to produce work in his papers. Keiichi Tokuda interconnects Adversarial system, Vibrato, Generative grammar and Deep neural networks in the investigation of issues within Singing voice synthesis.

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.

Best Publications

Statistical Parametric Speech Synthesis

A.W. Black;H. Zen;K. Tokuda.
international conference on acoustics, speech, and signal processing (2007)

1468 Citations

Speech parameter generation algorithms for HMM-based speech synthesis

K. Tokuda;T. Yoshimura;T. Masuko;T. Kobayashi.
international conference on acoustics, speech, and signal processing (2000)

1294 Citations

Voice Conversion Based on Maximum-Likelihood Estimation of Spectral Parameter Trajectory

T. Toda;A.W. Black;K. Tokuda.
IEEE Transactions on Audio, Speech, and Language Processing (2007)

1035 Citations

Simultaneous Modeling of Spectrum, Pitch and Duration in HMM-Based Speech Synthesis

Takayoshi Yoshimura;Keiichi Tokuda;Takashi Masuko;Takao Kobayashi.
conference of the international speech communication association (1999)

974 Citations

The HMM-based speech synthesis system (HTS) version 2.0.

Heiga Zen;Takashi Nose;Junichi Yamagishi;Shinji Sako.
SSW (2007)

635 Citations

A Speech Parameter Generation Algorithm Considering Global Variance for HMM-Based Speech Synthesis

Tomoki Toda;Keiichi Tokuda.
The IEICE transactions on information and systems (2007)

546 Citations

Speech parameter generation algorithm considering global variance for HMM-based speech synthesis

Tomoki Toda;Keiichi Tokuda.
conference of the international speech communication association (2005)

544 Citations

An adaptive algorithm for mel-cepstral analysis of speech

T. Fukada;K. Tokuda;T. Kobayashi;S. Imai.
international conference on acoustics, speech, and signal processing (1992)

490 Citations

AN HMM-BASED SPEECH SYNTHESIS SYSTEM APPLIED TO ENGLISH

Keiichi Tokuda;Heiga Zen;Alan W. Black.
(2003)

488 Citations

Speech Synthesis Based on Hidden Markov Models

K. Tokuda;Y. Nankaku;T. Toda;H. Zen.
Proceedings of the IEEE (2013)

473 Citations

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

Contact us

Best Scientists Citing Keiichi Tokuda

Junichi Yamagishi

Junichi Yamagishi

National Institute of Informatics

Publications: 191

Tomoki Toda

Tomoki Toda

Nagoya University

Publications: 153

Simon King

Simon King

University of Edinburgh

Publications: 106

Zhen-Hua Ling

Zhen-Hua Ling

University of Science and Technology of China

Publications: 100

Haizhou Li

Haizhou Li

Chinese University of Hong Kong, Shenzhen

Publications: 84

Satoshi Nakamura

Satoshi Nakamura

Nara Institute of Science and Technology

Publications: 84

Frank K. Soong

Frank K. Soong

Microsoft (United States)

Publications: 83

Thierry Dutoit

Thierry Dutoit

University of Mons

Publications: 61

Takao Kobayashi

Takao Kobayashi

Tokyo Institute of Technology

Publications: 56

Hiroshi Saruwatari

Hiroshi Saruwatari

University of Tokyo

Publications: 53

Hsin-Min Wang

Hsin-Min Wang

Academia Sinica

Publications: 52

Alan W. Black

Alan W. Black

Carnegie Mellon University

Publications: 52

Hirokazu Kameoka

Hirokazu Kameoka

NTT (Japan)

Publications: 46

Kiyohiro Shikano

Kiyohiro Shikano

Nara Institute of Science and Technology

Publications: 41

Paavo Alku

Paavo Alku

Aalto University

Publications: 40

Keikichi Hirose

Keikichi Hirose

University of Tokyo

Publications: 40

Something went wrong. Please try again later.