METHOD AND DEVICE FOR DECODING AUDIO SOUND FIELD REPRESENTATION FOR AUDIO REPRODUCTION

摘要

PROBLEM TO BE SOLVED: To provide a method and device for decoding audio sound field representation for audio reproduction.SOLUTION: A sound field signal like e.g., ambisonics carries representation of a desired sound field. An ambisonics format is based on a sphere harmonization function decomposition and high-order ambisonics (HOA) use at least a secondary sphere harmonization function. However, speaker setup used normally is irregular and a problem in decoder design may occur. A method for improved decoding of audio sound representation for audio reproduction includes a step (110) of using a geometrical method based on positions of a plurality of speakers and a plurality of source directions to compute a panning function (W); a step (120) of computing a mode matrix (&Xgr;) from the speaker positions; a step (130) of computing a pseudo inverse mode matrix (&Xgr;); and a step (130) of decoding the audio sound field representation. Decoding is based on a decode matrix D obtained from the panning function (W) and the pseudo inverse mode matrix (&Xgr;).