Lg Electronics Inc.

Lg Information & Communications, Ltd.

Alexandr L Maiboroda

Victor V Redkov

Andrei B Viktorov

Anatoli I Tikhotsky

Mikhael A Sall

Sergei N Gramnitskiy

Hyun Duk Cho

Anatoli Tikhotski

Eugene V Djourinski

Sung-Il Park

Mikhail Anatolyevich Sall

Hee-Suk Roh

Seong-Bin Lee

Valery Chernenko

Anatoli I Tikhotski

Sergei V Smolentsev

Chan-Hee Hong

Disclosed herein is a segmentation method, which divides an input audio stream into segments containing different homogeneous signals. The main objective of this method is localization of segments with stationary properties. This method seeks all no-stationary points or intervals in the audio stream and creates a list of segments. The obtained list of segments can be used as an input data for the following procedures, such as classification, speech/music/noise attribution and so on. The proposed segmentation method is based on the analysis of audio signal statistical features variation and comprises three main stages: stage of first-grade characteristics calculation, stage of second-grade characteristics calculation and stage of decision-making.

Method of segmenting an audio stream

A voice control apparatus of a vacuum cleaner includes: a sense signal detector for detecting a sense signal provided from a user to provide a voice recognizable environment in an ambience; a voice command recognizing unit for temporarily stopping a driving of a motor upon receiving the detected sense signal and, at the same time, outputting a voice command control signal previously set in a voice command set corresponding to a voice command instructed by the user; and a driving command controller for controlling a function and a mode of a vacuum cleaner upon receiving the voice command control signal. By providing a voice recognizable environment, the vacuum cleaner can be effectively controlled through a user's voice command even in an environment with a strong noise.

Voice control apparatus of vacuum cleaner and method thereof

A method of filtering digital audio data with short delay according to the present invention comprises the steps of: (a) buffering input source digital audio data; (b) calculating digital data that are being filtered from the buffered input source digital audio data; and (c) outputting a portion of the filtered digital data. The source digital audio data are sequence of digital samples. Buffer contents are shifted to arrange some of input samples in a buffer in the step (a). The step (b) comprises the steps of: (b-1) calculating a correlation matrix; (b-2) decomposing the correlation matrix; (b-3) calculating a filter matrix; and (b-4) calculating an approximation of the filtered digital data. In the step (b-2), the correlation matrix is represented by multiplication of three matrices where a first matrix is a matrix composed by normalized eigenvectors of the correlation matrix, a second matrix is a diagonal matrix in which eigenvalues of the correlation matrix are arranged on a principal diagonal and zeros are arranged in other positions, and a third matrix is a transposed matrix of the normalized eigenvectors of the correlation matrix.

Method of filtering noise of source digital data

The present invention relates to method and system for distinguishing speech from music in a digital audio signal in real time. A method for distinguishing speech from music in a digital audio signal in real time for the sound segments that have been segmented from an input signal of the digital sound processing systems by means of a segmentation unit on the base of homogeneity of their properties, comprises the steps of: (a) framing an input signal into sequence of overlapped frames by a windowing function; (b) calculating frame spectrum for every frame by FFT transform; (c) calculating segment harmony measure on base of frame spectrum sequence; (d) calculating segment noise measure on base of the frame spectrum sequence; (e) calculating segment tail measure on base of the frame spectrum sequence; (f) calculating segment drag out measure on base of the frame spectrum sequence; (g) calculating segment rhythm measure on base of the frame spectrum sequence; and (h) making the distinguishing decision based on characteristics calculated.

Method and system for distinguishing speech from music in a digital audio signal in real time

Method for encoding and decoding a picture signal for detecting and correcting errors in a new structured GOB according to a data partition technique, method for encoding a picture signal, including the steps of (1) grouping picture information of one block group into respective information regions of each block, and partitioning to relevant regions, and (2) forming a partition table having length information of the partitioned regions of the grouped respective information regions.

System and method for coding and decoding picture signals based on a partition table including length information

Device for encoding a DPCM image is provided. A smoothness test is performed on an original image at a smoothness tester. A linearly predicted value of the original image is obtained at a linear predictor. A switching part operates depending on a result of the smoothness test. A difference between the linearly predicted value and the original image is quantized, and encoded when the result of the smoothness test is satisfactory. A difference between the linearly predicted value corrected by using space self-likeness of the image and the original image is quantized, and encoded when the result of the smoothness test is not satisfactory. The adaptive application of the existing linear prediction, and the prediction corrected by using self-likeness of space likeness of the image permits to obtain a better predicted value for DPCM, and improve an image compression efficiency.

Device and method for encoding DPCM image

An apparatus and method for speech compression includes dividing the speech spectrum into a plurality of frames, assigning frame classifications to the plurality of frames, and determining the speech modeling parameters based on the assigned frame classification. The voiced part of the speech spectrum and the unvoiced part of the speech spectrum are synthesized separately using an Analysis by Synthesis allowing a correct correspondence between voiced and unvoiced parts of the reconstructed signal. Particularly, a frequency response of a special simulated signal based on the previous and current frames is used as an approximating function. The simulated signal is synthesized at the encoder side in the way it will be generated at the decoder side. Also, a better of two encoding methods is selected to encode the spectral magnitudes. A wavelet encoder and an inter-frame predictive encoder illustrate the invention's efficient, yet accurate reconstruction of synthesized digital speech.

Growing community of inventors

St. Petersburg, Russia

Alexandr L Maiboroda