【作者】 李晓明； 鲍长春； 贾懋珅；

【Author】 LI Xiao-ming BAO Chang-chun JIA Mao-shen (Speech and Audio Signal Processing Lab,School of Electronic Information and Control Engineering, Beijing University of Technology,Beijing 100124,China)

【机构】 北京工业大学电子信息与控制工程学院,语音与音频信号处理实验室；

【摘要】 针对MLT嵌入式编码器中需要确定不同频带的重要性,从而实现可分级编码的问题。本文依据人耳对于不同频率声音的感知特性,结合心理声学模型,提出了一种利用子带能量与掩蔽阈值的比值来确定其重要性的排序方法。通过对32 kHz采样信号进行实验,发现该方法在32/24/16 kbps码率下取得了较好的解码效果。本方法是心理声学模型应用于可分级编码的一种全新尝试。更多还原

【Abstract】 The MLT embedded audio encoder needs to determine the importance order of different frequency bands in order to achieve scalable coding.So this paper proposes a method of the sub-band importance ordering based on psychoacoustic model considering the perception of sound frequencies.The ratio between sub-band energy and the masking threshold is used as a key coefficient to determine the importance order.The test results for audio signal sampled at 32 kHz show that this method can achieve a good performance at the bit rates 32,24 and 16 kbps.This method is a novel attempt to use psychoacoustic model on scalable coding.更多还原