Learn More
In this paper, we propose an approach for combined acoustic echo and noise cancellation based on the GaussSeidel Pseudo Affine Projection algorithm (GS-PAP). It includes a simple residual echo cancellation scheme and a double-talk detector using a two-path model. Simulation results indicate that the proposed GS-PAP is stable, fast convergent and has good(More)
This paper presents an adaptive neuro-fuzzy filtering scheme using the artificial neuro-fuzzy inference system (ANFIS) for noise reduction in speech. The measurable output noisy speech with 5 dB SNR level is taken as the contaminated version of the interference to compare with the output data of the filter. The white noise source is taken as the input. With(More)
The application of a multilayer feedforward neural network to pulse radar detection or pulse compression is presented. For illustration, the Barker code was used. This network has 13 input units, 3 hidden units, and 1 output unit. Backpropagation learning was used to train the network. A 40-dB peak signal-to-noise ratio can be achieved easily. The(More)
A Levinson-type algorithm is proposed for estimating the two-dimensional autoregressive-moving average (ARMA) model from the observed inputs and outputs of an unknown 2-D system. By applying the concept of dummy prediction errors, the algorithm can be extended for computing 2-D models of any AR-order and MA-order. It is shown that the algorithm can be(More)
In this paper, a 2-D phoneme sequence pattern recognition using the fuzzy neural network is presented. The self-organizing map and the learning vector quantization are used to organize the phoneme feature vectors of short and long phonemes segmented from speech samples to obtain the phoneme maps. The 2-D phoneme response sequences of the speech samples are(More)