System for real time audio processing driven by interrupt and timer event

摘要

Provided are methods and systems for processing audio using a combination of interrupts and timer events. Buffer queues and timer events are configured and utilized between audio capture and encoding processes, and also between audio decoding and playout processes to deliver a fixed-rate audio packet and higher quality audio output. Thread execution priorities may be assigned to the capture and/or playout timer events, and such priorities may be automatically adjusted to conserve processing resources. By configuring timer events to drive audio encoding and decoding processes, and utilizing buffer queues to absorb variances in the timing of audio capture and playout interrupts, the methods and systems provided are able to deliver improved audio quality while conserving processing resources.

主权项

1. A computer-implemented method for processing audio data, the method comprising:
generating a capture interrupt event; in response to the capture interrupt event, copying audio data input from a capture device into an audio buffer configured to store the audio data according to one or more specified parameters; in response to a capture timer event, calculating an amount of audio data stored in the audio buffer, wherein the capture timer event is associated with a thread execution priority for encoding the audio data stored in the audio buffer; comparing the amount of audio data stored in the audio buffer with a detected latency of an operating system associated with the capture device; in response to the comparison indicating that the amount of audio data stored in the audio buffer satisfies a threshold, adjusting the thread execution priority for encoding the audio data stored in the audio buffer; and providing the audio data stored in the audio buffer to an encoder based on the thread execution priority or the adjusted thread execution priority.