TH1501004235A - Emphasizing the importance of low frequencies for LPC-based coding in the frequency domain. - Google Patents

Abstract

Edit Summary of Invention 10/21/2015 This invention is also provided. Audio codecs and methods for their encoding In order to generate a bit stream from an audio signal, the audio encoder is combined by a combination (2, 3) of a linear predictive coding filter (2) with a coefficient. Coded a linear prediction (LC) and time-frequency converter (3) where the aggregation (2, 3) is configured in order to be filtered and in order to convert the frame (FI) of The audio signal (AS) to the frequency domain to output a frame-based spectrum (SP) (FI) and on the coding coefficient. Linear projections (LC); The low-frequency priority (4) is configured to calculate the spectrum that Is processed (PS) that is based on the spectrum (SP), where the spectral line (SL) of the spectrum is Has been processed (PS) that represent frequencies lower than the reference spectral line (RSL) will be emphasized; And control devices (5) that have been configured to control the computation of the spectrum. That was processed (PS) by the low-frequency emphasis (4) that depends on the down coefficient. Linear Predictive Code (LC) of the Linear Predictive Coding Filter (2). Moreover, this invention is also prepared. Corresponding audio decoder, method system for decoding the contained bit stream. Quantized spectra and a number of linear predictive coding coefficients. And the corresponding computer program -------------------------------------------------- -------------------- This invention provides an audio encoder. And methods for encoding audio signals Non-speech in order to generate a bit stream from the audio signal. This audio encoder Also includes: a combination (2, 3) of filters for linear predictive coding (2) with coefficients. Predictive linear coding (LC) and time-frequency converter (3), where combinations (2, 3) are configured to filter. And to convert the frame (FI) of the audio signal (AS) into the frequency domain in order to output the spectrum (SP) based on the frame (FI) and on the down coefficient. Linear predictive code (plc); The low-frequency priority (4) that has been configured to calculate the processed spectrum (PS), which is spectral-based (SP), where the spectral line (S L) of the processed spectrum (PS) representing frequencies below the reference spectral line (RSL) are emphasized; And control devices (5) that have been configured to control the computations of the processed spectrum (PS) by the low-frequency emphasis (4) that are based on the coding coefficient. Filter linear prediction (LC) for linear predictive coding (2). Further, this invention provides audio decoders, systems, methods for the Decode the bit stream that contains the quantisated spectrum. And coding coefficient A number of linear predictions And a number of computer programs:

Claims (1)

: Edit Summary of Invention 10/21/2015 This invention is also provided. Audio codecs and methods for their encoding Non-speech audio signal In order to generate a bit stream from an audio signal, the audio encoder includes: a combination (2, 3) of a coefficient linear predictive coding filter (2). The Coded a linear prediction (LC) and time-frequency converter (3) where the aggregation (2, 3) is configured in order to be filtered and in order to convert the frame (FI) of The audio signal (AS) to the frequency domain to output a frame-based spectrum (SP) (FI) and on the coding coefficient. Linear projections (LC); The low-frequency priority (4) is configured to calculate the spectrum that Is processed (PS) that is based on the spectrum (SP), where the spectral line (SL) of the spectrum is Has been processed (PS) that represent frequencies lower than the reference spectral line (RSL) will be emphasized; And control devices (5) that have been configured to control the computation of the spectrum. That was processed (PS) by the low-frequency emphasis (4) that depends on the down coefficient. Linear Predictive Code (LC) of the Linear Predictive Coding Filter (2). Moreover, this invention is also prepared. Corresponding audio decoder, method system for decoding the contained bit stream. Quantized spectra and a number of linear predictive coding coefficients. And the corresponding computer program -------------------------------------------------- -------------------- This invention provides an audio encoder. And methods for encoding audio signals Non-speech in order to generate a bit stream from the audio signal. This audio encoder Also includes: a combination (2, 3) of filters for linear predictive coding (2) with coefficients. Predictive linear coding (LC) and time-frequency converter (3), where combinations (2, 3) are configured to filter. And to convert the frame (FI) of the audio signal (AS) into the frequency domain in order to output the spectrum (SP) based on the frame (FI) and on the down coefficient. Linear predictive code (plc); The low-frequency priority (4) that has been configured to calculate the processed spectrum (PS), which is spectral-based (SP), where the spectral line (S L) of the processed spectrum (PS) representing frequencies below the reference spectral line (RSL) are emphasized; And control devices (5) that have been configured to control the computations of the processed spectrum (PS) by the low-frequency emphasis (4) that are based on the coding coefficient. Filter linear prediction (LC) for linear predictive coding (2). Further, this invention provides audio decoders, systems, methods for the Decode the bit stream that contains the quantisated spectrum. And coding coefficient A number of linear predictions And a number of computer programs. (Section one) which will appear on the advertisement page: Amend 15/11/2018 Claims 1. An audio encoder for encoding a non-speech audio (AS) signal in order to Generates bits (BS) from audio signals. The audio encoder (1) consists of: a combination (2, 3) of the filter, coding, linear prediction (2) with coefficient. Coded a number of linear prediction (LC) and time-frequency plotter (2) where the aggregation (2,3) would be structured in order to filter and In order to drop the (FI) frame of the audio signal (As) to lock it by tag: