What is he best known for?
The fields of study he is best known for:
- Artificial intelligence
- Statistics
- Machine learning
His primary areas of investigation include Artificial intelligence, Speech recognition, Pattern recognition, Source separation and Recurrent neural network.
His Cluster analysis, Deep learning and Inference study, which is part of a larger body of work in Artificial intelligence, is frequently linked to Matrix decomposition, bridging the gap between disciplines.
His Speech recognition research is multidisciplinary, incorporating elements of Artificial neural network, Speech enhancement and Communication channel.
His Pattern recognition research includes themes of Channel and Monte Carlo method.
The Source separation study which covers Noise that intersects with Speech reconstruction.
As part of the same scientific family, John R. Hershey usually focuses on Recurrent neural network, concentrating on Applied mathematics and intersecting with Kullback–Leibler divergence, Divergence and Mixture model.
His most cited work include:
- Deep clustering: Discriminative embeddings for segmentation and separation (691 citations)
- Approximating the Kullback Leibler Divergence Between Gaussian Mixture Models (635 citations)
- Phase-sensitive and recognition-boosted speech separation using deep recurrent neural networks (402 citations)
What are the main themes of his work throughout his whole career to date?
John R. Hershey mostly deals with Speech recognition, Artificial intelligence, Pattern recognition, Speech enhancement and Artificial neural network.
His Speech processing, Language model, Hidden Markov model and Word error rate study in the realm of Speech recognition connects with subjects such as Sequence.
When carried out as part of a general Artificial intelligence research project, his work on Deep learning, Source separation and Discriminative model is frequently linked to work in Set, therefore connecting diverse disciplines of study.
John R. Hershey usually deals with Pattern recognition and limits it to topics linked to Cluster analysis and Spectrogram.
John R. Hershey focuses mostly in the field of Speech enhancement, narrowing it down to matters related to Algorithm and, in some cases, Masking.
His Artificial neural network research incorporates elements of End-to-end principle, Context and Decoding methods.
He most often published in these fields:
- Speech recognition (53.30%)
- Artificial intelligence (50.47%)
- Pattern recognition (27.83%)
What were the highlights of his more recent work (between 2019-2021)?
- Speech recognition (53.30%)
- Sound separation (4.25%)
- Artificial intelligence (50.47%)
In recent papers he was focusing on the following fields of study:
His scientific interests lie mostly in Speech recognition, Sound separation, Artificial intelligence, Sound and Separation.
John R. Hershey has included themes like Encoder, Feature, Encoding and Cluster analysis in his Speech recognition study.
His Sound separation research also works with subjects such as
- Sound sources which is related to area like Sound event detection,
- Open domain which connect with Algorithm.
The Artificial intelligence study combines topics in areas such as Signal-to-noise ratio, Beamforming and Pattern recognition.
The various areas that he examines in his Pattern recognition study include Artificial neural network, Speech enhancement, Covariance function and Word error rate.
His Sound research is multidisciplinary, relying on both Focus, Source separation and Benchmark.
Between 2019 and 2021, his most popular works were:
- Improving Universal Sound Separation Using Sound Classification (21 citations)
- Unsupervised Sound Separation Using Mixtures of Mixtures (18 citations)
- Sequential Multi-Frame Neural Beamforming for Speech Separation and Enhancement (9 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.
