Back to results
US 8,019,612 B2Granted
Methods for improving high frequency reconstruction
AssigneeCoding Technologies AB
InventorsKristofer Kjörling, Per Ekstrand, Holger Hörich
Life of this patent
Estimating the term…
Abstract
The present invention proposes a new method and a new apparatus for enhancement of audio source coding systems utilizing high frequency reconstruction (HFR). It utilizes a detection mechanism on the encoder side to assess what parts of the spectrum will not be correctly reproduced by the HFR method in the decoder. Information on this is efficiently coded and sent to the decoder, where it is combined with the output of the HFR unit.
Claims20
An encoder apparatus comprising:
- a coder configured to encode a lowband signal of an audio signal, the lowband signal being below a variable frequency;
- a high frequency regenerator configured to extrapolate a highband signal from the lowband signal, the highband signal including a component above the variable frequency;
- a detector configured to detect a difference between the highband signal and the audio signal above the variable frequency, the difference being a spectral line in the audio signal not included in the highband signal;
- and a describer configured to describe the difference as additional information, wherein the coder, high frequency regenerator, detector, and the describer comprise a hardware implementation.
1 more element
An encoder apparatus comprising:
- a coder encoding a lowband signal of an audio signal, the lowband signal being below a predetermined frequency;
- a high frequency regenerator extrapolating a highband signal from the lowband signal, the highband signal including a component above the predetermined frequency;
- a detector detecting a difference between the highband signal and the audio signal above the predetermined frequency, the difference being a spectral line in the audio signal not included in the highband signal;
- and a describer describing the difference as additional information, wherein the encoder apparatus varies the predetermined frequency, wherein the coder, high frequency regenerator, detector, and the describer comprise a hardware implementation.
1 more element
A method for encoding an audio signal, the method comprising:
- varying a predetermined frequency from a first frequency to a second frequency;
- encoding a lowband signal of the audio signal, the lowband signal being below the predetermined frequency;
- extrapolating a highband signal from the lowband signal, the highband signal including a component above the predetermined frequency;
- detecting a difference between the highband signal and the audio signal above the predetermined frequency, the difference being a spectral line in the audio signal not included in the highband signal;
- and
- describing the difference as additional information,
- wherein the method comprises a hardware implementation.
4 more elements
A decoder apparatus comprising:
- a decoder configured to decode an encoded input signal to produce a decoded input signal, the encoded input signal including a lowband signal of an audio signal and additional information, the lowband signal being below a variable frequency;
- a high frequency generator for performing a high frequency regeneration technique from the lowband signal to obtain a highband signal, the highband signal including a component above the variable frequency;
- a reconstructor configured to reconstruct a difference between the audio signal and the highband signal based on the additional information, wherein the difference is a spectral line in a specified frequency region, and the reconstructor is configured to generate a spectral line in the specified frequency region in response to the additional information;
- and a producer for producing a regenerated audio signal based on the lowband signal, the difference, and the highband signal, wherein the producer is arranged for determining band pass signals for filter bank channels in a scale factor band, and wherein the decoder, high frequency generator, reconstructor, and the producer comprise a hardware implementation.
1 more element
A non-transitory storage medium having stored thereon a computer program for performing the method for encoding an audio signal, the method comprising:
- varying a predetermined frequency from a first frequency to a second frequency;
- encoding a lowband signal of the audio signal, the lowband signal being below the predetermined frequency;
- extrapolating a highband signal from the lowband signal, the highband signal including a component above the predetermined frequency;
- detecting a difference between the highband signal and the audio signal above the predetermined frequency, the difference being a spectral line in the audio signal not included in the highband signal;
- and
- describing the difference as additional information.
3 more elements
Dependencies are inferred from the claim text.