Audio Decoder Multi-Band Noise Filling Bit Rate
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional noise filling techniques in audio coding often result in poor hearing impression due to low granularity or require excessive bit rate, as they either lack sufficient resolution or necessitate large amounts of side information.
Innovation Solution
A decoder that uses a single multi-band noise intensity value for noise filling, scaled by separate frequency band gain information, allowing for adaptive noise introduction based on psychoacoustic relevance without additional side information beyond what is required for audio content scaling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If conventional noise filling techniques are used with low granularity, then bit rate is reduced, but hearing impression degrades
Solution Approach 1:
The patent divides the frequency spectrum into multiple bands and applies different noise filling strategies to each band. By segmenting the spectrum, the system can use coarse noise filling in less critical bands (saving bits) while maintaining finer control in more critical bands (preserving hearing impression), thus resolving the contradiction between bit rate reduction and hearing quality.
Solution Approach 2:
The patent applies different noise filling granularities to different frequency bands based on their perceptual importance. Critical bands receive finer noise filling (better hearing impression) while non-critical bands receive coarser noise filling (lower bit rate), implementing local quality differentiation to resolve the contradiction between overall bit rate reduction and localized hearing quality.
2Object-affected harmful factors
If fine-grained noise filling is applied to improve hearing impression, then bit rate increases due to additional side information
Solution Approach 1:
The patent segments the frequency spectrum into multiple bands and transmits noise filling parameters selectively for each band. By segmentation, the system avoids transmitting excessive side information for all bands while maintaining fine-grained control where perceptually necessary, thus improving hearing impression without proportionally increasing bit rate.
Solution Approach 2:
The patent changes the parameter representation by using band-based noise filling control instead of spectral-line-based control. This parameter change reduces the total number of parameters that need to be transmitted, thereby improving hearing impression through better noise control while limiting the increase in bit rate.
3Quantity of substance
If spectral bins are quantized to zero to save bits, then bit rate is reduced, but audible artifacts are introduced
Solution Approach 1:
The patent converts the harmful effect of zero-quantized spectral bins (audible artifacts) into a beneficial noise filling opportunity. By introducing controlled noise in bands containing zero-quantized bins, the system masks the artifacts while maintaining the bit-saving advantage of zero quantization, thus resolving the contradiction between bit rate reduction and artifact introduction.
Solution Approach 2:
The patent introduces noise filling as an intermediary mechanism between the quantization process and the audible output. This intermediary noise masks the harmful artifacts generated by zero quantization while allowing the bit-saving quantization strategy to continue, thus resolving the contradiction between bit rate efficiency and artifact suppression.
Data Source
Figure 1
Figure 2
Figure 3A1
AI summary
An encoder for providing an audio stream on the basis of a transform-domain representation of an input audio signal comprises a quantization error calculator configured to determine a multi-band quantization error over a plurality of frequency bands of the input audio signal for which separate band gain information is available. The encoder also comprises an audio stream provider configured to provide the audio stream such that the audio stream comprises an information describing an audio content of the frequency bands and an information describing the multi-band quantization error. A decoder for providing a decoded representation of an audio signal on the basis of an encoded audio stream representing spectral components of frequency bands of the audio signal comprises a noise filler configured to introduce noise into spectral components of a plurality of frequency bands to which separate frequency band gain information is associated on the basis of a common multi-band noise intensity value.