Connectionist speech recognition of Broadcast News

By Anthony J. Robinson, Gary D. Cook, D. Ellis, Eric Fosler-Lussier, S. Renals, D. Williams. 2002

67 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Dong Yu
2. Li Deng
3. Jinyu Li
4. Yoshua Bengio
5. S. Renals
6. Tara N. Sainath
7. Andrew W. Senior
8. Yifan Gong

In Signal Processing

1. Dong Yu
2. Li Deng
3. Jinyu Li
4. S. Renals
5. Yifan Gong
6. Tara N. Sainath
7. Andrew W. Senior
8. Pawel Swietojanski

Citations in the corpus

• Regularization of context-dependent deep neural networks with context-independent multi-task training (2015, background): “The use of state-clustered context dependent (CD) phone units (also known as senones) was an key advance in the performance of HMM-GMM continuous speech recognition systems in the 1990s [1], giving an important competitive advantage over hybrid neural network based approaches such as [2, 3] used at ”
• Long short-term memory recurrent neural network architectures for large scale acoustic modeling (2014, methodology): “In acoustic modeling for speech recognition, however, where deep neural networks (DNNs) are the established state-of-the-art, recently RNNs have received little attention beyond small scale phone recognition tasks, notable exceptions being the work of Robinson [1], Graves [2], and Sak [3].”
• Multi-level adaptive networks in tandem and hybrid ASR systems (2013, background): “During the 1990s hybrid systems achieved good experimental results on some large vocabulary tasks [6, 7].”
• Context-Dependent Pre-Trained Deep Neural Networks for Large-Vocabulary Speech Recognition (2012, methodology): “Much previous hybrid ANN-HMM work focused on context-independent or rudimentary context-dependent phone models and small to mid-vo-cabulary tasks (with notable exceptions such as [45]), possibly masking some of the potential advantages of the ANN-HMM hybrid approach.”
• Tandem Connectionist Feature Extraction for Conversational Speech Recognition (2004, methodology): “This inherently discriminant approach worked well for many tasks, and was particularly useful for combinations of features with different statistical properties (e.g., continuous and binary features) [ 13 ].”
• Discriminative Input Stream Combination for Conditional Random Field Phone Recognition (2009, methodology): “aging models for combining posterior estimators [4], [5].”