What is he best known for?
The fields of study he is best known for:
- Artificial intelligence
- Speech recognition
- Statistics
His primary scientific interests are in Speech recognition, Transcription, Word error rate, Artificial intelligence and NIST.
In his works, Thomas Hain performs multidisciplinary study on Speech recognition and Component.
His Transcription study deals with Segmentation intersecting with Cluster analysis and Audio mining.
The Word error rate study combines topics in areas such as Artificial neural network, Feature, Speaker recognition and Speech processing.
As a part of the same scientific family, Thomas Hain mostly works in the field of Artificial intelligence, focusing on Natural language processing and, on occasion, Gesture and Context.
Thomas Hain studied NIST and Vocal tract that intersect with Discriminative model.
His most cited work include:
- The AMI meeting corpus: a pre-announcement (549 citations)
- The AMI meeting corpus (301 citations)
- The AMI System for the Transcription of Speech in Meetings (117 citations)
What are the main themes of his work throughout his whole career to date?
Thomas Hain mostly deals with Speech recognition, Artificial intelligence, Transcription, Natural language processing and Word error rate.
His Speaker recognition, NIST, Hidden Markov model, Acoustic model and Speech processing study are his primary interests in Speech recognition.
His research investigates the connection between Artificial intelligence and topics such as Pattern recognition that intersect with problems in Feature.
In his research, Speech technology is intimately related to Multimedia, which falls under the overarching field of Transcription.
His Natural language processing research includes elements of Speech corpus and Phone.
His study explores the link between Word error rate and topics such as Test set that cross with problems in Latent Dirichlet allocation.
He most often published in these fields:
- Speech recognition (72.14%)
- Artificial intelligence (34.83%)
- Transcription (22.89%)
What were the highlights of his more recent work (between 2018-2021)?
- Speech recognition (72.14%)
- Encoder (4.98%)
- Embedding (4.98%)
In recent papers he was focusing on the following fields of study:
His primary areas of investigation include Speech recognition, Encoder, Embedding, TIMIT and Speaker recognition.
Thomas Hain has included themes like Context, Adaptation and Robustness in his Speech recognition study.
His research in Encoder intersects with topics in Segmentation and Speaker identification.
His TIMIT research is multidisciplinary, relying on both Pooling and Gaussian noise.
His Speaker recognition study combines topics in areas such as Speech enhancement and Joint.
His Utterance study combines topics from a wide range of disciplines, such as NIST, Representation, American English and Discriminative model.
Between 2018 and 2021, his most popular works were:
- Learning Temporal Clusters Using Capsule Routing for Speech Emotion Recognition. (10 citations)
- Recurrent Neural Network Language Model Adaptation for Multi-Genre Broadcast Speech Recognition and Alignment (9 citations)
- Unsupervised Acoustic Unit Representation Learning for Voice Conversion Using WaveNet Auto-Encoders. (7 citations)
In his most recent research, the most cited papers focused on:
- Artificial intelligence
- Statistics
- Speech recognition
Thomas Hain focuses on Speech recognition, Emotion recognition, Recurrent neural network, Speaker recognition and Word error rate.
Thomas Hain has researched Speech recognition in several fields, including Context, Embedding, Encoder, Representation and Convolutional neural network.
His Embedding research integrates issues from NIST, American English and Discriminative model.
His Emotion recognition research is included under the broader classification of Artificial intelligence.
The concepts of his Speaker recognition study are interwoven with issues in Speech enhancement, Frequency domain, Attention model and Robustness.
His Word error rate research includes themes of Language model, Perplexity, Feature, Adaptation and Test set.
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.
