Sound Image Localization for Remote Voice Distinction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing remote conferencing techniques fail to distinguish between the voice of a real participant and a remote participant, as they do not account for the presence of real persons or objects in the space, leading to confusion in voice localization.
Innovation Solution
An information processing apparatus that localizes the sound image of a remote participant's voice to a position different from that of real participants using a sound image localization processing unit, which sets localization positions within localizable regions excluding the real participants' positions, allowing for clear differentiation between voices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the voice of a remote participant is localized to a position in the predetermined space, then the remote participant can be visually located in the space, but confusion arises when the localized position coincides with the position of a real participant
Solution Approach 1:
The patent applies local quality by making the localization position specific to each remote participant based on their participant ID. Each remote participant is assigned a unique localization position within the predetermined space, allowing their voice to be localized to a distinct location that does not coincide with real participants or other remote participants, thereby maintaining both localization accuracy and participant identity distinction
Solution Approach 2:
The patent introduces an intermediary mechanism (the sound image localization processing unit) that mediates between the remote participant's voice and its localized position. This intermediary processes the voice signal and places it at an appropriate position in the predetermined space, preventing direct confusion with real participants while maintaining the remote participant's presence in the virtual space
2Ease of operation
If virtual speech positions are set at intervals in front of the listener, then remote voices can be localized, but it becomes difficult to distinguish when a real person is present at that position
Solution Approach 1:
The patent applies preliminary action by determining the positions of real participants before localizing remote participant voices. The sound image localization processing unit first identifies where real participants are located, then selects virtual speech positions that do not coincide with these real participant positions, thereby preventing confusion before it occurs
Solution Approach 2:
The patent implements dynamics by making the virtual speech positions adjustable and adaptable. When real participants move or when the conference environment changes, the system can dynamically reassign virtual speech positions to maintain distinction between real and remote participants, ensuring continuous reliability of voice distinction
Data Source
AI summary
The present technique relates to an information processing apparatus, an information processing method, and a program that make it easy to distinguish between the voice of a real participant and the voice of a remote participant.An information processing apparatus according to one aspect of the present technique includes a sound image localization processing unit that localizes a sound image of a voice of a remote participant, who is participating remotely in a conversation conducted in a predetermined space, to a position different from a position of a real participant who is a participant present in the predetermined space. The present technique can be applied in computers which perform remote conferencing.