Computationally efficient audio coder

摘要

The present invention provides a computationally efficient technique for compression encoding of an audio signal, and further provides a technique to enhance the sound quality of the encoded audio signal. This is accomplished by including more accurate attack detection and a computationally efficient quantization technique. The improved audio coder converts the input audio signal to a digital audio signal. The audio coder then divides the digital audio signal into larger frames having a long-block frame length and partitions each of the frames into multiple short-blocks. The audio coder then computes short-block audio signal characteristics for each of the partitioned short-blocks based on changes in the input audio signal. The audio coder further compares the computed short-block characteristics to a set of threshold values to detect presence of an attack in each of the short-blocks and changes the long-block frame length of one or more short-blocks upon detecting the attack in the respective one or more short-blocks.

主权项

1. A method of reducing computation during quantization iterations for compression of an input audio signal to improve the efficiency of operation of a perceptual encoder, comprising:
initializing a common_scalefac value of a current frame; initializing a quantizer_change value of the current frame; computing counted_bits associated with the current frame; computing a difference between the computed counted_bits and available_bits; comparing the computed difference with the a pre-determined MAXDIFF value; if the computed difference is greater than the pre-determined MAXDIFF value, then resetting the common_scalefac value and quantizer_change value; and if the computed difference is less than or equal to the pre-determined MAXDIFF value, then changing common_scalefac value based on the outcome of the comparison; wherein resetting the common_scalefac value and the quantizer_change value further comprises: computing predicted_common_scalefac value based on stored common_scalefac value of the previous frame adjacent to the current frame; if counted_bits is greater than available_bits and if the start_common_scalefac value+64 is not greater than predicted_common_scalefac value, then resetting the common_scalefac value to the start_common_scalefac value+64; if the counted_bits is less than or equal to available_bits and the common_scalefac is not greater than start_common_scalefac+32, then the common_scalefac is set to start_common_scalefac+32, and the quantizer_change is set to 32 and counted_bits is recomputed if predicted_common_scalefac is greater than common_scalefac; and if the counted_bits is less than or equal to available_bits, and the common_scalefac value is greater than the start_common_scalefac+32, then the common_scalefac is set to start_common_scalefac+64 and counted bits is recomputed if predicted_common_scalefac is greater than common_scalefac.