3.3 Framework for Audio Pre-Processing

The current framework is based on DSP/BIOS. The following are the overall steps:

• The I2S/DMA is configured to generate a DMA Completion Interrupt every 10 ms.
• The DMA ISR prepares the audio frame buffer pointers for further processing.
• The DMA ISR also sends a semaphore to wake up the main task.
• When woke up, the main task performs the BF, ASNR, MSS and DRC.
• The main task also outputs the final processing audio frame and the unprocessed audio input to the I2S/DMA TX buffers.
• After completed, the audio frame processing, the main thread will wait on semaphore for next audio frames to come.