High-performance robust speech recognition using stereo training data

By Li Deng, A. Acero, Li Jiang, Jasha Droppo, Xuedong Huang. 2001

155 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, 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. Li Deng
2. Dong Yu
3. Jinyu Li
4. Yoshua Bengio
5. Jianfeng Gao
6. Yifan Gong
7. Quoc V. Le
8. Arun Narayanan

In Computer Vision and Pattern Recognition

1. Li Deng
2. Jianfeng Gao
3. Dong Yu
4. Luc Van Gool
5. Trevor Darrell
6. Dacheng Tao
7. Xiaogang Wang
8. Ming-Hsuan Yang

In Signal Processing

1. Li Deng
2. Dong Yu
3. Jinyu Li
4. Yifan Gong
5. Arun Narayanan
6. Deliang Wang
7. Shinji Watanabe
8. Heiga Zen

Citations in the corpus

• Deep Learning: Methods and Applications (2014, methodology): “See a comprehensive review in [220] and some additional review literature or original work in [4, 82, 119, 140, 230, 370, 404, 431, 444].”
• A fast maximum likelihood nonlinear feature transformation method for GMM-HMM speaker adaptation (2014, background): “The second class of techniques [7,8] adapts the model or feature by learning the distortion model from stereo data, i.”
• Voice conversion in high-order eigen space using deep belief nets (2013, methodology): “The VC techniques have been applied to various tasks, such as speech enhancement [1], emotion conversion [2], speaking assistance [3], and other applications [4, 5].”
• Distributed speech processing in miPad's multimodal user interface (2002, methodology): “The basic SPLICE optimally processes each frame of noisy speech independently.”
• An Overview of Noise-Robust Automatic Speech Recognition (2014, methodology): “In reality, when and are Gaussians given component , their joint distribution can be modeled as (61) and a rotation on is needed for the conditional mean as (62) where (63) (64) The feature compensation formulation in this case is (65) It is interesting that feature space minimum phone error (fMPE) ”
• MiPad: a multimodal interaction prototype (2001, background): “Our noise robustness code has been improved to deal to improve the performance of the built -in microphone under both seen and unseen conditions [ 6 ,7].”
• Continuous Stochastic Feature Mapping Based on Trajectory HMMs (2011, background): “…speech to air-transmitted speech conversion [20], [36], [49], narrowband to wideband speech transformation [7], [24], [29], articulatory-to-acoustic mapping [21], [30], [33], acoustic-to-articulatory inversion mapping [10], [27], [35], speaking aids [14], [22], and noise compensation [2], [4], [5].”
• Recurrent Neural Networks for Noise Reduction in Robust ASR (2012, methodology): “Training noise reduction models using stereo (clean and noisy) data has been used successfully with the SPLICE algorithm [6].”
• Investigation of Speech Separation as a Front-End for Noise Robust Speech Recognition (2014, methodology): “When stereo data is available, feature mapping methods like SPLICE [7] and recurrent neural networks [25] have been used.”