Audio coding and decoding method and apparatus

Abstract

The invention discloses an audio encoding method and a corresponding decoding method, which help transient audio signals to avoid preliminary echoes and minimize the distortion of the transient audio signals. The invention adopts the technical proposal that: the transient signals inputted into the audio are processed with time domain treatment; the sampling points x1, x2,,,xN of the inputted frame are divided into L segments; wherein N is the length of the inputted frame and L is any natural number less than N; the energy Ei of each segment is calculated, wherein i is a natural number from 1 to L; the average energy E0 of the segments of the inputted frame is calculated; the multiplicative coefficient corresponding to each segment is calculated: (Lambada)i = r(bitrate)*E0 / Ei, wherein i is a natural number from 1 to L, and r(bitrate) is a function correlative with bit rate; sampling points of all the segments of the inputted frame are multiplied by corresponding multiplicative coefficients (Lambada)i to obtain the processed sampling points x1`, x2`,,,Xn`, and meanwhile, the multiplicative coefficients (Lambada)i are sent to a bit stream for transmission; the processed sampling points x1`, x2`,,,Xn` undergo time frequency transform coding and then are outputted to the bit stream. The audio encoding method and corresponding decoding method are applicable in mobile communication field.

Description

technical field [0001] The present invention relates to a coding and decoding method and device, in particular to a method and device for coding and decoding audio signals. Background technique [0002] A transient signal is a special audio signal that mostly exists in an audio sequence with percussion instruments. For example, the signal generated by continuous beating of gongs and drums can be called a transient signal. Its particularity lies in that if it is encoded by conventional transform coding, such as MDCT (Modified Discrete Cosine Transform) method, pre-echo phenomenon will occur due to the existence of quantization noise. The reason for the pre-echo phenomenon is due to the quantization noise caused by insufficient quantization bits. The quantization noise spreads evenly throughout the time domain. The signal before the transient signal will be occupied by the quantization noise, thereby generating the pre-echo phenomenon. . The pre-echo phenomenon is a kind of ...

AI Technical Summary

Problems solved by technology

Switching between long and short windows requires a lot of computing overhead and takes up a lot of buffer space. The time-domain noise shaping method uses the results of adaptive prediction in the frequency domain to shape the distribution of quantization noise in the time domain. The processing method is relatively simple. However, due to its incomplete extraction of the time-domain envelope, it will produce some other distortions

Images