Method and system for split client-server reverberation processing

摘要

In some embodiments, a method applying reverberation to audio from at least one client of a set of clients which share a virtual environment, including by asserting position data and at least one input audio stream to a server, selecting (in the server) a reverberation filter for each input audio stream in response to the position data and generating wet audio by applying to the input audio an early reverberation part of the selected reverberation filter. Typically, a client applies a late reverberation filter to the wet audio using metadata from the server. In other embodiments, a server selects a reverberation filter for application to audio in response to position data, asserts the audio and metadata indicative of the filter, and a client applies the filter to the audio using the metadata. Other aspects are systems, servers, and client devices configured to perform any embodiment of the method.

主权项

1. A method for applying reverberation to audio from at least one client of a set of clients which share a virtual environment, said method including steps of:
(a) asserting position data and at least one input audio stream to a server, wherein each said input audio stream is indicative of audio from a different one of the clients, the position data is indicative of a listener position and at least one source position in the virtual environment, and each said source position is a position in the virtual environment of a source of one said input audio stream; (b) in response to the position data, operating the server to select a reverberation filter for each said input audio stream; and (c) generating a wet audio stream for each said input audio stream, by operating the server to apply to the input audio stream an early reverberation part but not a late reverberation part of the reverberation filter selected for said input audio stream.