A hybrid Maxent/HMM based ASR system
نویسندگان
چکیده
The aim of this work is to develop a practical framework, which extends the classical Hidden Markov Models (HMM) for continuous speech recognition based on the Maximum Entropy (MaxEnt) principle. The MaxEnt models can estimate the posterior probabilities directly as with Hybrid NN/HMM connectionist speech recognition systems. In particular, a new acoustic modelling based on discriminative MaxEnt models is formulated and is being developed to replace the generative Gaussian Mixture Models (GMM) commonly used to model acoustic variability. Initial experimental results using the TIMIT phone task are reported.
منابع مشابه
A Hybrid MaxEnt/HMM ba
The aim of this work is to develop a practical framework, which extends the classical Hidden Markov Models (HMM) for continuous speech recognition based on the Maximum Entropy (MaxEnt) principle. The MaxEnt models can estimate the posterior probabilities directly as with Hybrid NN/HMM connectionist speech recognition systems. In particular, a new acoustic modelling based on discriminative MaxEn...
متن کاملA Hybrid Oriya Named Entity Recognition system: Harnessing the Power of Rule
This paper describes a hybrid system that applies maximum entropy (MaxEnt) model with Hidden Markov model (HMM) and some linguistic rules to recognize name entities in Oriya language. The main advantage of our system is, we are using both HMM and MaxEnt model successively with some manually developed linguistic rules. First we are using MaxEnt to identify name entities in Oria corpus, then tagg...
متن کاملHybrid HMM/Neural Network based Speech Recognition in Loquendo ASR
This paper describes hybrid Hidden Markov Models / Artificial Neural Networks (HMM/ANN) models devoted to speech recognition, and in particular Loquendo HMM/ANN, that is the core of Loquendo ASR. While Hidden Markov Models (HMM) is a dominant approach in most state-of-the-art speaker-independent, continuous speech recognition systems (and commercial products), Artificial Neural Networks (ANN) a...
متن کاملOn recognition of non-native speech using probabilistic lexical model
Despite various advances in automatic speech recognition (ASR) technology, recognition of speech uttered by non-native speakers is still a challenging problem. In this paper, we investigate the role of different factors such as type of lexical model and choice of acoustic units in recognition of speech uttered by non-native speakers. More precisely, we investigate the influence of the probabili...
متن کاملHybrid SVM/HMM model for the arab phonemes recognition
Hidden Markov Models (HMM) are currently widely used in Automatic Speech Recognition (ASR) as being the most effective models. Yet, they sometimes pose some problems of discrimination. The hybridization of Artificial Neural Networks (ANN) in particular Multi Layer Perceptrons (MLP) with HMM is a promising technique to overcome these limitations. In order to ameliorate results of recognition sys...
متن کامل