USAC Decoder FFT and LPC Decoding With Precomputed Lookup Tables
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Unified Audio and Speech (USAC) encoders and decoders require significant computational resources due to complex computation steps, leading to high processing loads during runtime.
Innovation Solution
The implementation of a core decoder for USAC streams that includes an upmixing unit with a decorrelator using pre-computed filter coefficients, a QMF based harmonic transposer, and a fast Fourier transform module based on the Cooley-Tuckey algorithm, reducing computational load by referencing pre-computed values for filter coefficients and twiddle factors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If complex computation steps are used in USAC encoders and decoders, then decoding functionality is achieved, but computational load during runtime increases
Solution Approach 1:
The patent pre-computes filter coefficients for the decorrelation filter and stores them in a lookup table before runtime. During decoding, the apparatus simply retrieves pre-computed coefficients based on the channel configuration, avoiding complex real-time calculations while maintaining accurate decorrelation functionality.
Solution Approach 2:
The patent pre-computes twiddle factors for the FFT module and stores them in a lookup table. During runtime, the FFT module retrieves these pre-computed values, significantly reducing the computational complexity of the Fast Fourier Transform operations while preserving transform accuracy.
2Productivity
If pre-computed values are used for filter coefficients, then computational load is reduced, but memory requirements increase
Solution Approach 1:
The patent organizes pre-computed filter coefficients in a lookup table indexed by channel configuration parameters. This allows the system to store coefficients for multiple configurations efficiently and retrieve only the necessary coefficients based on the current channel configuration, optimizing the balance between memory usage and computational efficiency.
Solution Approach 2:
The patent pre-computes and stores only the essential filter coefficients and twiddle factors that are actually needed for different channel configurations, rather than storing all possible values. This selective pre-computation approach minimizes memory requirements while still providing the computational benefits of pre-computed values.
Data Source
AI summary
The present disclosure relates to an apparatus for decoding an encoded Unified Audio and Speech stream. The apparatus comprises a core decoder for decoding the encoded Unified Audio and Speech stream. The core decoder includes a fast Fourier transform, FFT, module implementation based on a Cooley-Tuckey algorithm. The FFT module is configured to determine a discrete Fourier transform, DFT. Determining the DFT involves recursively breaking down the DFT into small FFTs based on the Cooley-Tucker algorithm and using radix-4 if a number of points of the FFT is a power of 4 and using mixed radix if the number is not a power of 4. Performing the small FFTs involves applying twiddle factors. Applying the twiddle factors involves referring to pre-computed values for the twiddle factors. The present disclosure further relates to an apparatus for decoding an encoded Unified Audio and Speech stream, in which the core decoder is configured to decode an LPC filter that has been quantized using a line spectral frequency, LSF, representation from the Unified Audio and Speech stream. Decoding the LPC filter from the Unified Audio and Speech stream comprises computing a first-stage approximation of a LSF vector, reconstructing a residual LSF vector, if an absolute quantization mode has been used for quantizing the LPC filter, determining inverse LSF weights for inverse weighting of the residual LSF vector by referring to pre-computed values for the inverse LSF weights or their respective corresponding LSF weights, inverse weighting the residual LSF vector by the determined inverse LSF weights, and calculating the LPC filter based on the inversely-weighted residual LSF vector and the first-stage approximation of the LSF vector. The present disclosure further relates to corresponding methods and storage media.