TDNN labeling for a HMM recognizer

Ma, Weiye; Compernolle, Dirk Van

doi:10.1109/icassp.1990.115728

Cited by 9 publications

(2 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Strictly related to the idea of using ANNs as vector quantizers for discrete HMMs is the concept of neural labelers [77,17]. In its basic version, this system is based on an MLP (or TDNN), fed with input acoustic features, with an output unit per each phonetic class.…”

Section: Network As Vector Quantizers For Discrete Hmmmentioning

confidence: 99%

“…It is a modular combination of TDNNs, where the nets are trained as phoneme classi"ers by BP on a training set labeled by hand. TDNNs internal representations (activations of the second hidden layer) of the acoustic features are used as inputs for a discrete HMM, the latter being separately trained using standard algorithms SI ("ve males for training, two males for test), isolated Korean words (75 words dictionary) recognition task: 44.9% relative WER reduction with respect to the standard discrete HMM Neural labelers [77,17] An MLP (or TDNN) with an output unit per each phonetic class is trained to yield output values equal to 0 or 1, according to the membership of the input observation to the class or not. The output from the MLP is passed on to a standard discrete HMM that uses it as a class label within a Viterbi decoding paradigm.…”

Section: Modelmentioning

confidence: 99%

See 1 more Smart Citation

A survey of hybrid ANN/HMM models for automatic speech recognition

2001

View full text Add to dashboard Cite

In spite of the advances accomplished throughout the last decades, automatic speech recognition (ASR) is still a challenging and di$cult task. In particular, recognition systems based on hidden Markov models (HMMs) are e!ective under many circumstances, but do su!er from some major limitations that limit applicability of ASR technology in real-world environments. Attempts were made to overcome these limitations with the adoption of arti"cial neural networks (ANN) as an alternative paradigm for ASR, but ANN were unsuccessful in dealing with long time-sequences of speech signals. Between the end of the 1980s and the beginning of the 1990s, some researchers began exploring a new research area, by combining HMMs and ANNs within a single, hybrid architecture. The goal in hybrid systems for ASR is to take advantage from the properties of both HMMs and ANNs, improving #exibility and recognition performance. A variety of di!erent architectures and novel training algorithms have been proposed in literature. This paper reviews a number of signi"cant hybrid models for ASR, putting together approaches and techniques from a highly specialistic and non-homogeneous literature. E!orts concentrate on describing and referencing architectures and algorithms, their advantages and limitations, as well as on categorizing them into broad classes. Early attempts to emulate HMMs by ANNs are "rst described. Then we focus on ANNs to estimate posterior probabilities of the states of an HMM and on`globala optimization, where a single, overall training criterion is de"ned over the HMM and the ANNs. Connectionist vector quantization for discrete HMMs, and other more recent approaches are also reviewed. It is pointed out that, in addition to their theoretical interest, hybrid systems have been allowing for tangible improvements in recognition performance over the standard HMMs in di$cult and signi"cant benchmark tasks.

show abstract

Section: Network As Vector Quantizers For Discrete Hmmmentioning

confidence: 99%

Section: Modelmentioning

confidence: 99%

A survey of hybrid ANN/HMM models for automatic speech recognition

2001

View full text Add to dashboard Cite

show abstract

Toward the border between neural and Markovian paradigms

Wilinski¹,

Solaiman

Hillion

et al. 1998

IEEE Trans. Syst., Man, Cybern. B

View full text Add to dashboard Cite

A new tendency in the design of modern signal processing methods is the creation of hybrid algorithms. This paper gives an overview of different signal processing algorithms situated halfway between Markovian and neural paradigms. A new systematic way to classify these algorithms is proposed. Four specific classes of models are described. The first one is made up of algorithms based upon either one of the two paradigms, but including some parts of the other one. The second class includes algorithms proposing a parallel or sequential cooperation of two independent Markovian and neural parts. The third class tends to show Markov models (MMs) as a special case of neural networks (NNs), or conversely NNs as a special case of MMs. These algorithms concentrate mainly on bringing together respective learning methods. The fourth class of algorithms are hybrids, neither purely Markovian nor neural. They can be seen as belonging to a more general class of models, presenting features from both paradigms. The first two classes essentially include models with structural modifications, while two later classes propose algorithmic modifications. For the sake of clarity, only main mathematical formulas are given. Specific applications are intentionally avoided to give a wider view of the subject. The references provide more details for interested readers.

show abstract

Maximum mutual information neural networks for hybrid connectionist-HMM speech recognition systems

Rigoll

1994

IEEE Trans. Speech Audio Process.

View full text Add to dashboard Cite

This paper proposes a novel approach for a hybrid connectionist-hidden Markov model (HMM) speech recognition system based on the use of a neural network as vector quantizer. The neural network is trained with a new learning algorithm offering the following innovations:1) It is an unsupervised learning algorithm for perceptronlike neural networks that are usually trained in supervised mode. 2) Information theory principles are used as learning criteria, making the network especially suitable for combination with a HMM-based speech recognition system. 3) The neural network is not trained using the standard error-backpropagation algorithm but using instead a newly developed self-organizing learning approach. The use of the hybrid system with the neural vector quantizer results in a 25% error reduction compared with the same HMM system using a standard k-means vector quantizer. The training algorithm can be further refined by using a combination of unsupervised and supervised learning algorithms. Finally, it is demonstrated how the new learning approach can be applied to multiple-feature hybrid speech recognition systems, using a joint information theory-based optimization procedure for the multiple neural codebooks, resulting in a 30% error reduction.

show abstract

TDNN labeling for a HMM recognizer

Cited by 9 publications

References 4 publications

A survey of hybrid ANN/HMM models for automatic speech recognition

A survey of hybrid ANN/HMM models for automatic speech recognition

Toward the border between neural and Markovian paradigms

Maximum mutual information neural networks for hybrid connectionist-HMM speech recognition systems

Contact Info

Product

Resources

About