Transitional speech units and their representation by regressive Markov states: applications to speech recognition

By Li Deng, Hossein Sameti. 1996

How researchers see this paper

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

1. Li Deng
2. Dong Yu
3. Li Deng
4. Yoshua Bengio
5. Jinyu Li
6. Abdel-rahman Mohamed
7. Geoffrey E. Hinton
8. Jianfeng Gao

Citations in the corpus

• Deep Learning: Methods and Applications (2014, background): “Specifically, the early work described in [71, 72, 83, 84, 99, 274] generalized and extended the conventional shallow and conditionally independent HMM structure by imposing dynamic constraints, in the form of polynomial trajectory, on the HMM parameters.”
• Dynamic Speech Models: Theory, Algorithms, and Applications (2006, methodology): “, [50, 65]).”
• An Overview of Deep-Structured Learning for Information Processing (2011, unknown): “Specifically, the early work described in (Deng 1992, 1993; Deng et al., 1994; Ostendorf et al., 1996, Deng and Sameti, 1996) generalized and extended the conventional shallow HMM structure by imposing dynamic constraints, in the form of polynomial trajectory, on the HMM parameters.”
• Computational Models for Speech Production (2018, background): “Within the class of acoustic-dynamic models, the model which attempts to condition the properties of the dynamic directly on specific feature-coded speech production mechanisms is described in [7].”
• An overlapping-feature-based phonological model incorporating linguistic constraints: applications to speech recognition. (2002, methodology): “The next version of the model improved the phonological rules for constraining feature overlaps, and interfaced the feature-bundles with the HMM states which are nonstationary (polynomial) [8, 6].”
• Graphical model architectures for speech recognition (2005, background): “Examples include an audio and a video feature stream [2], [21], [22], [39], [40], differing streams of audio features [48], different articulatory sequences [ 17 ], [19], [30], [33], [34], [43], or different spectral subband streams [14], [23], [35].”
• Switching Dynamic System Models for Speech Articulation and Acoustics (2004, methodology): “We refer the readers to some earlier work for ways of constructing such over­ lapping units, either by rules or by automatic learning, which have proved effective in the HMM-like speech recognition framework [21, 19 , 18, 56] .”