SYSTEM FOR GENERATING IMMERSIVE AUDIO UTILIZING VISUAL CUES

摘要

The present disclosure is directed to a system for generating immersive audio utilizing visual cues. In general, a system may be capable of transforming non-spatial audio (e.g., simple mono or stereo sound) associated with video into immersive sound (e.g., wherein sound may be generated spatially so that it appears to emanate directly from sound sources in the video). An example device may comprise data sourcing circuitry to receive multimedia data including at least video and non-spatial audio corresponding to the video, data analysis circuitry to determine at least one source of audio in the video and at least one attributable sound in the audio and audio generation circuitry to generate immersive audio wherein the at least one attributable sound is at least spatially associated with the at least one source of audio. Audio sources in the video may be determined using a variety of different types of detection.

主权项

1. At least one device for generating immersive audio, comprising:
data analysis circuitry to analyze multimedia data including video and non-spatial audio, wherein analyzing the multimedia data includes determining at least one source of audio within the video and at least one attributable sound within the non-spatial audio; and audio generation circuitry to generate immersive audio wherein the at least one attributable sound is at least spatially associated with the at least one source of audio.