Speech Recognition Using Augmented Conditional Random Fields

By Yasser Hifny, S. Renals. 2009

95 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

In Signal Processing

Citations in the corpus

Deep Learning: Methods and Applications (2014, background): “Many of the discriminative techniques for supervised learning in signal and information processing are shallow architectures such as HMMs [52, 127, 147, 186, 188, 290, 394, 418] and conditional random fields (CRFs) [151, 155, 281, 400, 429, 446].”
A tutorial survey of architectures, algorithms, and applications for deep learning (2014, background): “This difficulty is partially responsible for steering away most of the machine learning and signal-processing research from neural networks to shallow models that have convex loss functions (e.g., SVMs, CRFs, and MaxEnt models), for which global optimum can be efficiently obtained at the cost of les”
Context-Dependent Pre-Trained Deep Neural Networks for Large-Vocabulary Speech Recognition (2012, methodology): “2134090 fields (CRFs) [18]–[20], hidden CRFs [21], [22], and segmental CRFs [23]).”
Phone recognition using Restricted Boltzmann Machines (2010, background): “Table 1 shows the phone error rate (PER) when the RBMs were trained generatively.”
Backpropagation training for multilayer conditional random field based phone recognition (2010, methodology): “This is especially true in ASR, where the input speech is usually represented by MFCC or PLP coefficients, and previous studies using CRFs for phone recognition [2] [3] [4] have demonstrated the importance of suitably transforming this input representation to derive feature functions for CRFs.”
Equivalence of Generative and Log-Linear Models (2011, methodology): “The normalization constant is computed by running over all tag strings of length .”
TIMIT Acoustic-Phonetic Continuous Speech Corpus (2012, background): “In early 2009, Hifny and Renals (Hifny & Renals, 2009) presented a phonetic recognition system where the acoustic modulation is achieved by means of augmented conditional random fields.”
Deep Belief Networks for phone recognition (2009, methodology): “Many methods have been proposed for relaxing the very strong conditional independence assumptions of standard HMMs (e.g. [1] [2] [3] [4]).”
Phone Recognition on the TIMIT Database (2012, background): “In early 2009, Hifny and Renals (Hifny & Renals, 2009) presented a phonetic recognition system where the acoustic modulation is achieved by means of augmented conditional random fields.”
Machine Learning Paradigms for Speech Recognition: An Overview (2013, methodology): “For example, when applying (26) to Markov random fields, we obtain the training objective of conditional random fields (CRFs) [70]: (27) The partition function is a normalization factor. is a weight vector and is a vector of feature functions referred to as a feature vector.”
An Overview of Deep-Structured Learning for Information Processing (2011, background): “Many of the discriminative techniques in signal and information processing apply to shallow architectures such as HMMs (e.g., He et al., 2008; Jiang and Li, 2010; Xiao and Deng, 2010; Gibson and Hain, 2010) or CRFs (e.g., Yang and Furui, 2009; Yu, Li, and Deng, 2010; Hifny and Renals, 2009).”
On Speech Recognition (2012, methodology): “Hidefumi et al. (1982a) developed a low cost speaker-dependent speech recognition unit using WalshHadamard transform (WHT). A WHT LSI has been developed to reduce the cost and the space of the recognition unit and a high rate of recognition has been obtained. The speech recognition algorithm and the”
Segmental Recurrent Neural Networks for End-to-End Speech Recognition (2016, background): “, [2, 3, 4, 5]; however these approaches have not yet improved speech recognition accuracy over HMMs.”
Acoustic Modeling Using Deep Belief Networks (2012, unknown): “8% Large-Margin GMM [33] 33% CD-HMM [34] 27.”
Phone Recognition with the Mean-Covariance Restricted Boltzmann Machine (2010, unknown): “3% Augmented conditional Random Fields [20] 26.”
Deep Neural Networks for Acoustic Modeling in Speech Recognition (2012, unknown): “AugMenteD ConDitionAl RAnDoM FielDs [26] 26.”
Learning a better representation of speech soundwaves using restricted boltzmann machines (2011, unknown): “Method PER Large-Margin GMM[14] 30.1% CD-HMM[15] 27.3% Augmented Conditional Random Fields[15] 26.6% Recurrent Neural Nets[16] 26.1% Bayesian Triphone HMM[17] 25.6% Monophone HTMs[18] 24.8% Heterogenous Classifiers[19] 24.4% Deep Belief Networks (bounding silences ignored)[9] DBN using mean covarian”