Add to ORKGAbstract: Robust speaker-independent alphabet recognition over a telephone line is a task yet unsolved. The current study examines the feasibility of Modular Recurrent Neural Networks (MRNNs) successfully applied to Mandarin syllable recognition to the recognition of German letters. Letters were divided into sub-word units for each of which specialized RNNs were trained. The results presented in this paper show that, within the framework of MRNNs, the highest recognition rates can be achieved by segmenting letters into right context-dependent initials and context-independent finals. However, these rates are lower than those achieved with comparable HMMs, primarily due to the higher robustness against segmentation inaccuracies of the latter ...
An artificial neural network has been trained by the error back-propagation technique to recognise p...
Neural networks have recently been applied to real-world speech recognition problems with a great de...
In this paper we present our results on using RNN-based LM scores trained on different phone-gram or...
A new modular recurrent neural network (MRNN)- based speech-recognition method that can recognize th...
Recurrent neural network language models (RNNLMs) have been successfully applied in a variety of lan...
In this paper, we investigate phone sequence modeling with recurrent neural networks in the context ...
It has been shown through a number of experiments that neural networks can be used for a phonetic ty...
A recognition system is reported which recognizes names spelled over the telephone with brief pauses...
This paper presents a distinctive phonetic features (DPFs) based phoneme recognition method by incor...
[[abstract]]A real time Mandarin speech recognition method using the classification technique of hyp...
This paper presents a phoneme recognition method based on distinctive phonetic features (DPFs). The ...
Abstract. The paper investigates the use of Singular and Modular Neural Networks in classify-ing the...
. Here we report about investigations concerning the application of Fully Recurrent Neural Networks ...
This paper describes phone modelling improvements t o the hybrid ronnectionist-hidden Markov model s...
Recurrent neural networks (RNNs) are a powerful model for sequential data. End-to-end training metho...
An artificial neural network has been trained by the error back-propagation technique to recognise p...
Neural networks have recently been applied to real-world speech recognition problems with a great de...
In this paper we present our results on using RNN-based LM scores trained on different phone-gram or...
A new modular recurrent neural network (MRNN)- based speech-recognition method that can recognize th...
Recurrent neural network language models (RNNLMs) have been successfully applied in a variety of lan...
In this paper, we investigate phone sequence modeling with recurrent neural networks in the context ...
It has been shown through a number of experiments that neural networks can be used for a phonetic ty...
A recognition system is reported which recognizes names spelled over the telephone with brief pauses...
This paper presents a distinctive phonetic features (DPFs) based phoneme recognition method by incor...
[[abstract]]A real time Mandarin speech recognition method using the classification technique of hyp...
This paper presents a phoneme recognition method based on distinctive phonetic features (DPFs). The ...
Abstract. The paper investigates the use of Singular and Modular Neural Networks in classify-ing the...
. Here we report about investigations concerning the application of Fully Recurrent Neural Networks ...
This paper describes phone modelling improvements t o the hybrid ronnectionist-hidden Markov model s...
Recurrent neural networks (RNNs) are a powerful model for sequential data. End-to-end training metho...
An artificial neural network has been trained by the error back-propagation technique to recognise p...
Neural networks have recently been applied to real-world speech recognition problems with a great de...
In this paper we present our results on using RNN-based LM scores trained on different phone-gram or...