Layered audio coding and transmission

摘要

Embodiments of systems and methods are described for generating layered audio such that computing devices can request a variable amount of data based on criteria such as their available bandwidth, device capability, or user selection. A base layer and one or more enhancement layers that incrementally enhance the previous layers may be generated. A computing device may retrieve the base layer and/or one or more enhancement layers, adjusting, in real-time or near real-time, which layers are retrieved based on fluctuations in the available bandwidth among other possible criteria.

主权项

1. A method of generating layered audio, the method comprising:
accessing a first audio frame and a second audio frame; comparing the first audio frame with the second audio frame; identifying a match between audio data in a first portion of the first audio frame and audio data in a second portion of the second audio frame based on the comparison; substituting the audio data in the second portion in the second audio frame with a reference to a location in the first audio frame that corresponds with the audio data in the first portion to create a modified second audio frame in response to identifying the match; generating a first audio layer based on the first audio frame; generating a second audio layer based on the modified second audio frame; and making the first audio layer and the second audio layer available for transmission to a client device over a network, wherein transmission of the first audio layer over the network requires a first amount of network bandwidth and transmission of both the first audio layer and the second audio layer over the network requires a second amount of network bandwidth that is greater than the first amount of network bandwidth, enabling the client device to receive and output the first audio layer and the second audio layer together if the second amount of network bandwidth is available to the client device while still enabling the client device to retrieve and output the first audio layer if the first amount of network bandwidth is available to the client device.