# US 2023/0410822 A1 · Filling of Non-Coded Sub-Vectors in Transform Coded Audio Signals

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2023/0410822 A1
- Document ID: US20230410822A1
- Status: Published application
- Kind: A1
- Application number: 18/365,322
- Filed: 2023-08-04
- Published: 2023-12-21
- Earliest claimed: 2011-03-10
- Assignees: Telefonaktiebolaget LM Ericsson (publ)
- Applicants: Telefonaktiebolaget LM Ericsson (publ)
- Inventors: Volodya Grancharov; Sebastian Näslund; Sigurdur Sverrisson
- Web page: https://patentgrep.com/patent/US20230410822A1

## CPC

- G10L 19/02: using spectral analysis, e.g. transform vocoders or subband vocoders
- G10L 19/0212: {using orthogonal transformation}
- G10L 19/028: Noise substitution, i.e. substituting non-tonal spectral components by noisy source (comfort noise for discontinuous speech transmission G10L19/012)
- G10L 19/038: Vector quantisation, e.g. TwinVQ audio
- G10L 21/038: using band spreading techniques
- G10L 2019/0007: {Codebook element generation}
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING

## Abstract

A spectrum filler for filling non-coded residual sub-vectors of a transform coded audio signal includes a sub-vector compressor configured to compress actually coded residual sub-vectors. A sub-vector rejecter is configured to reject compressed residual sub-vectors that do not fulfill a predetermined sparseness criterion. A sub-vector collector is configured to concatenate the remaining compressed residual sub-vectors to form a first virtual codebook. A coefficient combiner is configured to combine pairs of coefficients of the first virtual codebook to form a second virtual codebook. A sub-vector filler is configured to fill non-coded residual sub-vectors below a predetermined frequency with coefficients from the first virtual codebook, and to fill non-coded residual sub-vectors above the predetermined frequency with coefficients from the second virtual codebook.

## Claims

### Claim 1

A method of audio decoding, the method comprising:
receiving an encoded transform vector comprising coded residual sub-vectors corresponding to respective frequency bands of a time-domain frame of an audio signal, wherein coded residual sub-vectors for one or more frequency bands are missing;
decoding the coded residual sub-vectors to obtain corresponding decoded residual sub-vectors;
forming a first codebook from coefficients of the decoded residual sub-vectors and forming a second codebook from frequency-mirror pairs from the first codebook;
for each missing coded residual sub-vector, generating a corresponding residual sub-vector using coefficients from the first or second codebook, in dependence on whether the corresponding frequency band is above or below a cutoff frequency; and
reconstructing the transform vector from the decoded residual sub-vectors and the one o more generated residual sub-vectors.
