Other

Industry-University Cooperation Foundation Sogang University

Mpwav Inc.

Sogang University Research Business Development Foundation

Byung Joon Cho

Hyung Min Park

Seoyoung Lee

Uihyeop Shin

Seung-Yun Kim

A target signal extraction apparatus according to an embodiment of the present invention may comprise a steering vector estimator and a beamformer. The steering vector estimator may generate an input signal covariance according to input results for each frequency over time, generate a noise covariance on the basis of a variance determined according to output results corresponding to the input results, and generate a steering vector on the basis of the input signal covariance and the noise covariance. The beamformer may generate a beamforming weight according to a beamforming covariance determined according to the variance and the steering vector, and provide the output results on the basis of the input results and the beamforming weight. The target signal extraction apparatus according to the present invention may generate the steering vector by calculating the noise covariance on the basis of the variance determined according to output results corresponding to input results, and increases extraction performance for a target sound source by updating a beamforming weight.

Beamforming method using online likelihood maximization combined with steering vector estimation for robust speech recognition, and apparatus therefor

A method of noise cancellation comprises providing a first estimated output signal based on an input signal and a first prior far-end channel filter and a second prior far-end channel filter corresponding to a prior frame, updating a first current far-end channel filter corresponding to a current frame according to the first estimated output signal and the first prior far-end channel filter, providing a second estimated output signal based on the input signal, the first current far-end channel filter, and the second prior far-end channel filter, updating a second current far-end channel filter corresponding to the current frame according to the second estimated output signal and the second prior far-end channel filter, and providing a resultant signal based on the input signal, the first current far-end channel filter, and the second current far-end channel filter.

Device and method for stereo noise cancellation

Provided is a target speech signal extraction method for robust speech recognition including: receiving information on a direction of arrival of the target speech source with respect to the microphones; generating a nullformer by using the information on the direction of arrival of the target speech source to remove the target speech signal from the input signals and to estimate noise; setting a real output of the target speech source using an adaptive vector as a first channel and setting a dummy output by the nullformer as a remaining channel; setting a cost function for minimizing dependency between the real output of the target speech source and the dummy output using the nullformer by performing independent component analysis (ICA) or independent vector analysis (IVA); setting an auxiliary function to the cost function; and estimating the target speech signal by using the cost function and the auxiliary function.

Online target-speech extraction method based on auxiliary function for robust automatic speech recognition

According to an embodiment, a noise cancellation method comprises updating a first far-end channel filter according to a first variance yielded based on an input signal and a first far-end signal included in the input signal, providing a first output signal obtained by removing a first far-end noise signal from the input signal through the first far-end channel filter, updating a second far-end channel filter according to a second variance yielded based on the first output signal and a second far-end signal included in the input signal, and providing a second output signal obtained by removing a second far-end noise signal from the first output signal through the second far-end channel filter.

Growing community of inventors

Seoul, South Korea

Byung Joon Cho