Speech-Centric Information Processing: An Optimization-Oriented Approach

By Xiaodong He, Li Deng. 2013

49 citations recorded by the source corpus.

Subfields: Artificial Intelligence

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. Jianfeng Gao
4. Yoshua Bengio
5. Jinyu Li
6. Xiaodong He
7. James R. Glass
8. Hung-yi Lee

Citations in the corpus

• From Eliza to XiaoIce: challenges and opportunities with social chatbots (2018, background): “components are jointly optimized to cope with the large variability occurring in dialog systems (He and Deng, 2013; Sarikaya et al., 2016; Wen et al., 2016).”
• Deep Reinforcement Learning: An Overview (2017, background): “…al., 2016); semi-supervised learning (Zhu and Goldberg, 2009); Bayesian RL (Ghavamzadeh et al., 2015); spoken dialogue systems (Hinton et al., 2012; He and Deng, 2013; Young et al., 2013); AI safety (Amodei et al., 2016; Garc`ıa and Fern`andez, 2015), Monte Carlo tree search (MCTS) (Browne et…”
• Spoken Content Retrieval—Beyond Cascading Speech Recognition with Text Retrieval (2015, methodology): “performance of the respective application tasks is better than optimizing the ASR module and the downstream processing separately [105].”
• Using Recurrent Neural Networks for Slot Filling in Spoken Language Understanding (2015, background): “Further, the recurrent neural networks for the SLU application as presented in this paper can be generalized to other types of speech-centric information processing tasks [67], [68].”
• An Overview of Noise-Robust Automatic Speech Recognition (2014, background): “…with mobile devices (e.g., Siri on iPhone, Bing voice search on winPhone, and Google Now on Android), voice control in home entertainment systems (e.g., Kinect on xBox), and various speech-centric information processing applications capitalizing on downstream processing of ASR outputs [3].”
• Optimization Algorithms and Applications for Speech and Language Processing (2013, methodology): “However, many optimization problems in speech and language processing are nonconvex in nature, due to the complexity of the models (e.g., having latent variables) in use and to the fact that loss functions are designed to measure actual recognition errors [10], [22].”