| 摘要 |
A signal processing method, comprising: acquiring voice signals of at least two channels and acquiring frequency domain audio signals corresponding to the voice signal of each channel; acquiring a beam-forming output signal of a beam group corresponding to an audio signal of each frequency point; acquiring an output direction of the beam group; acquiring an outputted time domain voice signal after the beam is formed in said direction. A signal processing device, adopting a broadband beam-forming algorithm based on the frequency domain, and effectively improving a gain in voice reception; the device adaptively selects the optimal beam, avoiding the need to provide prior information such as the direction of arrival of an expected signal, thus reducing algorithm complexity and expanding the scope of application of the algorithm. The adopted frequency domain beam-forming algorithm facilitates fine adjustment of a signal frequency spectrum, thereby facilitating integration with other pre- and post-processing algorithms, while also being easy to implement with little calculation and suitable for all types of embedded platforms. |
