Encoder, decoder and methods for signal-adaptive switching of the overlap ratio in audio transform coding

An encoder for encoding a plurality of time-domain audio samples of an audio signal by generating a plurality of groups of spectral domain-audio samples from a plurality of groups of the time-domain audio samples, wherein the encoder comprises:

• a first encoding module for generating a first group of the groups of spectral-domain audio samples from a first group of the groups of the time-domain audio samples, and for generating a second group of the groups of spectral-domain audio samples from a second group of the groups of the time-domain audio samples, wherein the first group of time-domain audio samples and the second group of time-domain audio samples are neighboured in time within the groups of the time-domain audio samples, wherein the first group of the time-domain audio samples comprises more than 5% and at most 50% of the audio samples of the second group of the time-domain audio samples, and wherein the second group of the time-domain audio samples comprises more than 5% and at most 50% of the audio samples of the first group of the time-domain audio samples, and
• a second encoding module for generating a third group of the groups of spectral-domain audio samples from a third group of the groups of the time-domain audio samples, and for generating a fourth group of the groups of spectral-domain audio samples from a fourth group of the groups of the time-domain audio samples, wherein the third group of the time-domain audio samples comprises more than 60% and less than 100% of the audio samples of the fourth group of the time-domain audio samples, and wherein the fourth group of the time-domain audio samples comprises more than 60% and less than 100% of the audio samples of the third group of the time-domain audio samples, and
• an output module for outputting the first group of spectral-domain audio samples, the second group of spectral-domain audio samples, the third group of spectral-domain audio samples and the fourth group of spectral-domain audio samples,
• wherein the third group of the time-domain audio samples comprises audio samples of the second group of the time-domain audio samples, or wherein the fourth group of the time-domain audio samples comprises audio samples of the first group of the time-domain audio samples.

A method for decoding a plurality of spectral-domain audio samples, wherein the method comprises:

• decoding a first group of the spectral-domain audio samples by generating a first group of time-domain intermediate audio samples from the spectral-domain audio samples of the first group of the spectral-domain audio samples, and decoding a second group of the spectral-domain audio samples by generating a second group of time-domain intermediate audio samples from the spectral-domain audio samples of the second group of the spectral-domain audio samples,
• overlap-adding of exactly two groups of time-domain intermediate audio samples, said exactly two groups being the first group and the second group of time-domain intermediate audio samples, wherein said exactly two groups are overlap-added with an overlap of more than 5% and at most 50%, wherein said overlap-add of said exactly two groups results in generating a first plurality of time-domain audio output samples of an audio signal,
• decoding a third group of the spectral-domain audio samples by generating a third group of time-domain intermediate audio samples from the spectral-domain audio samples of the third group of the spectral-domain audio samples, and decoding a fourth group of the spectral-domain audio samples by generating a fourth group of time-domain intermediate audio samples from the spectral-domain audio samples of the fourth group of the spectral-domain audio samples,
• outputting the first plurality of time-domain audio output samples of the audio signal, a second plurality of time-domain audio output samples of the audio signal and a third plurality of time-domain audio output samples of the audio signal,
• acquiring the second plurality of time-domain audio output samples using overlap-adding at least the third group of time-domain intermediate audio samples with an overlap of more than 60% and less than 100% with the fourth group of time-domain intermediate audio samples, and
• acquiring the third plurality of time-domain audio output samples using overlap-adding of at least the second group of time-domain intermediate audio samples with the third group of time-domain intermediate audio samples, or acquiring the third plurality of time-domain audio output samples using overlap-adding of at least the fourth group of time-domain intermediate audio samples with the first group of time-domain intermediate audio samples.

A method for encoding a plurality of time-domain audio samples of an audio signal by generating a plurality of groups of spectral domain-audio samples from a plurality of groups of the time-domain audio samples, wherein the encoder comprises:

• generating a first group of the groups of spectral-domain audio samples from a first group of the groups of the time-domain audio samples, and generating a second group of the groups of spectral-domain audio samples from a second group of the groups of the time-domain audio samples, wherein the first group of time-domain audio samples and the second group of time-domain audio samples are neighboured in time within the groups of the time-domain audio samples, wherein the first group of the time-domain audio samples comprises more than 5% and at most 50% of the audio samples of the second group of the time-domain audio samples, and wherein the second group of the time-domain audio samples comprises more than 5% and at most 50% of the audio samples of the first group of the time-domain audio samples,
• generating a third group of the groups of spectral-domain audio samples from a third group of the groups of the time-domain audio samples, and generating a fourth group of the groups of spectral-domain audio samples from a fourth group of the groups of the time-domain audio samples, wherein the third group of the time-domain audio samples comprises more than 60% and less than 100% of the audio samples of the fourth group of the time-domain audio samples, and wherein the fourth group of the time-domain audio samples comprises more than 60% and less than 100% of the audio samples of the third group of the time-domain audio samples, and
• outputting the first group of spectral-domain audio samples, the second group of spectral-domain audio samples, the third group of spectral-domain audio samples and the fourth group of spectral-domain audio samples,
• wherein the third group of the time-domain audio samples comprises audio samples of the second group of the time-domain audio samples, or wherein the fourth group of the time-domain audio samples comprises audio samples of the first group of the time-domain audio samples.

A non-transitory digital storage medium having a computer program stored thereon to perform the method for decoding a plurality of spectral-domain audio samples, wherein the method comprises:

• decoding a first group of the spectral-domain audio samples by generating a first group of time-domain intermediate audio samples from the spectral-domain audio samples of the first group of the spectral-domain audio samples, and decoding a second group of the spectral-domain audio samples by generating a second group of time-domain intermediate audio samples from the spectral-domain audio samples of the second group of the spectral-domain audio samples,
• overlap-adding of exactly two groups of time-domain intermediate audio samples, said exactly two groups being the first group and the second group of time-domain intermediate audio samples, wherein said exactly two groups are overlap-added with an overlap of more than 5% and at most 50%, wherein said overlap-add of said exactly two groups results in generating a first plurality of time-domain audio output samples of an audio signal,
• decoding a third group of the spectral-domain audio samples by generating a third group of time-domain intermediate audio samples from the spectral-domain audio samples of the third group of the spectral-domain audio samples, and decoding a fourth group of the spectral-domain audio samples by generating a fourth group of time-domain intermediate audio samples from the spectral-domain audio samples of the fourth group of the spectral-domain audio samples,
• outputting the first plurality of time-domain audio output samples of the audio signal, a second plurality of time-domain audio output samples of the audio signal and a third plurality of time-domain audio output samples of the audio signal,
• acquiring the second plurality of time-domain audio output samples using overlap-adding at least the third group of time-domain intermediate audio samples with an overlap of more than 60% and less than 100% with the fourth group of time-domain intermediate audio samples, and
• acquiring the third plurality of time-domain audio output samples using overlap-adding of at least the second group of time-domain intermediate audio samples with the third group of time-domain intermediate audio samples, or acquiring the third plurality of time-domain audio output samples using overlap-adding of at least the fourth group of time-domain intermediate audio samples with the first group of time-domain intermediate audio samples,
• when said computer program is run by a computer.

A non-transitory digital storage medium having a computer program stored thereon to perform the method for encoding a plurality of time-domain audio samples of an audio signal by generating a plurality of groups of spectral domain-audio samples from a plurality of groups of the time-domain audio samples, wherein the encoder comprises:

• generating a first group of the groups of spectral-domain audio samples from a first group of the groups of the time-domain audio samples, and generating a second group of the groups of spectral-domain audio samples from a second group of the groups of the time-domain audio samples, wherein the first group of time-domain audio samples and the second group of time-domain audio samples are neighboured in time within the groups of the time-domain audio samples, wherein the first group of the time-domain audio samples comprises more than 5% and at most 50% of the audio samples of the second group of the time-domain audio samples, and wherein the second group of the time-domain audio samples comprises more than 5% and at most 50% of the audio samples of the first group of the time-domain audio samples,
• generating a third group of the groups of spectral-domain audio samples from a third group of the groups of the time-domain audio samples, and generating a fourth group of the groups of spectral-domain audio samples from a fourth group of the groups of the time-domain audio samples, wherein the third group of the time-domain audio samples comprises more than 60% and less than 100% of the audio samples of the fourth group of the time-domain audio samples, and wherein the fourth group of the time-domain audio samples comprises more than 60% and less than 100% of the audio samples of the third group of the time-domain audio samples, and outputting the first group of spectral-domain audio samples, the second group of spectral-domain audio samples, the third group of spectral-domain audio samples and the fourth group of spectral-domain audio samples,
• wherein the third group of the time-domain audio samples comprises audio samples of the second group of the time-domain audio samples, or wherein the fourth group of the time-domain audio samples comprises audio samples of the first group of the time-domain audio samples,
• when said computer program is run by a computer.