Audio stream processing method and audio stream processing device

The invention provides an audio stream processing method and an audio stream processing device, which are used to solve a distortion problem of a prior art of a way of highlighting a voice in audio frequency. The audio stream processing method comprises steps that energy of a frequency component in a preset frequency range of an input first audio stream is enhanced to acquire a second audio stream; energy of the background voice of the second audio stream is reduced to acquire a third audio stream; the third audio stream is output. The audio frequency is processed by adopting the audio stream processing method, the voice distortion is small, and an effect of highlighting a human voice is obvious. 本发明提供了种音频流的处理方法及装置，用于解决现有技术中在音频中突出语音的方式存在失真的问题，该方法包括：增强输入的第音频流中预设频率范围内的频率分量的能量，得到第二音频流；降低第二音频流中的背景声音的能量，得到第三音频流；输出第三音频流，使用该方法对音频进行处理，使得语音的失真较小，突显人声的效果较为明显。