What is he best known for?
The fields of study he is best known for:
- Artificial intelligence
- Speech recognition
- Statistics
Speech recognition, Speech enhancement, Speech processing, Reverberation and Artificial neural network are his primary areas of study.
His biological study spans a wide range of topics, including Mixture model and Noise reduction.
He interconnects Source separation and Blind signal separation in the investigation of issues within Mixture model.
His work deals with themes such as Time domain and Deep neural networks, which intersect with Speech enhancement.
His Speech processing research incorporates themes from Speaker recognition, Noise measurement and Robustness.
The Reverberation study combines topics in areas such as Filter, Microphone, Linear prediction, Impulse response and Signal processing.
His most cited work include:
- Speech Dereverberation Based on Variance-Normalized Delayed Linear Prediction (214 citations)
- A summary of the REVERB challenge: state-of-the-art and remaining challenges in reverberant speech processing research (186 citations)
- Making Machines Understand Us in Reverberant Rooms: Robustness Against Reverberation for Automatic Speech Recognition (182 citations)
What are the main themes of his work throughout his whole career to date?
Tomohiro Nakatani spends much of his time researching Speech recognition, Artificial intelligence, Speech enhancement, Pattern recognition and Reverberation.
His study in Speech recognition is interdisciplinary in nature, drawing from both Artificial neural network and Noise.
His study looks at the intersection of Artificial intelligence and topics like Blind signal separation with Expectation–maximization algorithm and Underdetermined system.
The study incorporates disciplines such as Microphone array, Noise reduction, Background noise and Beamforming in addition to Speech enhancement.
Tomohiro Nakatani has included themes like Noise measurement and Word error rate in his Pattern recognition study.
His Reverberation research integrates issues from Linear prediction, Filter, Microphone and Inverse filter.
He most often published in these fields:
- Speech recognition (64.69%)
- Artificial intelligence (32.08%)
- Speech enhancement (24.26%)
What were the highlights of his more recent work (between 2018-2021)?
- Speech recognition (64.69%)
- Artificial neural network (16.98%)
- Algorithm (15.09%)
In recent papers he was focusing on the following fields of study:
Tomohiro Nakatani focuses on Speech recognition, Artificial neural network, Algorithm, Speech enhancement and Artificial intelligence.
The various areas that Tomohiro Nakatani examines in his Speech recognition study include Time domain, Estimator and Beamforming.
His Artificial neural network research incorporates themes from Language model, Discriminative model, Utterance and Word error rate.
His Algorithm study combines topics in areas such as Noise, Filter and Blind signal separation.
Tomohiro Nakatani interconnects Noise reduction, Filter bank, Reverberation and Minimum-variance unbiased estimator in the investigation of issues within Speech enhancement.
His study focuses on the intersection of Artificial intelligence and fields such as Pattern recognition with connections in the field of Neural network learning and Learning methods.
Between 2018 and 2021, his most popular works were:
- Improving transformer-based end-to-end speech recognition with connectionist temporal classification and language model integration (60 citations)
- All-neural Online Source Separation, Counting, and Diarization for Meeting Analysis (31 citations)
- A Unified Convolutional Beamformer for Simultaneous Denoising and Dereverberation (27 citations)
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.
