METHOD AND DEVICE TO ESTIMATE MIXTURE RATIO OF VOICE AND MUSIC AND AUDIO DEVICE USING THE SAME

摘要

PROBLEM TO BE SOLVED: To effectively emphasize narration sound of audio signals so as to easily listen to the sound. SOLUTION: In the mixture ratio estimating method of voice and music, feature data of voice and music signals are beforehand learned and stored using pure narration voice and music. Then, the mixture ratio of the narration voice and the music of the audio signals is estimated using the stored feature data and the feature data of audio signals being broadcasted.