Unlock instant, AI-driven research and patent intelligence for your innovation.

Audio encoding/decoding based on efficient representation of autoregressive coefficients

An autoregressive coefficient and coding technology, applied in speech analysis, instruments, etc., can solve the problems of undesigned quantization scheme, large signal perception error, etc., and achieve the effect of reducing computational complexity

Active Publication Date: 2016-09-14
TELEFON AB LM ERICSSON (PUBL)
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, conventional quantization schemes are not designed for the case when the AR coefficients model treble frequencies (e.g., above 6 kHz) and operate with a very limited bit budget (which does not allow for transparent coding of the coefficients).
This introduces large perceptual errors in the reconstructed signal when using these traditional quantization schemes in non-optimal frequency ranges and at non-optimal bitrates

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Audio encoding/decoding based on efficient representation of autoregressive coefficients
  • Audio encoding/decoding based on efficient representation of autoregressive coefficients
  • Audio encoding/decoding based on efficient representation of autoregressive coefficients

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] The proposed technique requires as input a vector a of AR coefficients (another common name is Linear Prediction (LP) coefficients). These coefficients are usually obtained by first computing the autocorrelation r(j) of the windowed audio segment s(n) (n=l,...,N), namely:

[0036] r ( j ) = Σ n = j N s ( n ) s ( n - j ) , j = 0 , . . . , M - - - ( 1 )

[0037] where M is the predefined model order. Then, the AR coefficient a is obtained according to the autocorrelation sequence r(j) through...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

An encoder (50) for encoding a parametric spectral representation (f) partially representing autoregressive coefficients of an audio signal is described. The encoder comprises a low frequency encoder (10) configured to quantize elements of a portion of the parametric spectral representation corresponding to a low frequency portion of the audio signal. The encoder also includes a high frequency encoder (12) configured to pass the quantized elements (fL) flipped based on the quantized image frequency (fm) centered and during the closed loop search The high frequency part (fH) of the parametric spectral representation (f) is encoded by weighted averaging of the frequency grid determined from the frequency grid codebook (24), where the quantized image frequency (fm) combines the low frequency part with the high frequency The frequency part is separated. Corresponding decoders, corresponding encoding / decoding methods and UEs comprising such encoders / decoders are also described.

Description

technical field [0001] The proposed technique involves audio encoding / decoding based on efficient representation of autoregressive (AR) coefficients. Background technique [0002] AR analysis is commonly used in time-domain audio coding [1] and transform-domain audio coding [2]. Different applications use AR vectors with different lengths (the model order mainly depends on the bandwidth of the encoded signal; from 10 coefficients for a signal with a bandwidth of 4 kHz to 24 coefficients for a signal with a bandwidth of 16 kHz). These AR coefficients are quantized using separate multi-level vector quantization (VQ), which ensures nearly transparent reconstruction. However, conventional quantization schemes are not designed for the case when the AR coefficients model treble frequencies (eg, above 6 kHz) and operate with a very limited bit budget (which does not allow transparent coding of the coefficients). This introduces large perceptual errors in the reconstructed signal ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G10L19/06G10L21/038G10L19/032G10L19/02
CPCG10L19/06G10L19/0204G10L21/038G10L19/032G10L19/038G10L2019/001
Inventor 沃洛佳·格兰恰诺夫西德格尔·思维里森
Owner TELEFON AB LM ERICSSON (PUBL)