Dynamically preventing audio artifacts
The disclosure is directed to a process that can predict and prevent an audio artifact from occurring. The process can monitor the systems, processes, and execution threads on a larger system/device, such as a mobile or in-vehicle device. Using a learning algorithm, such as deep neural network (DNN), the information collected can generate a prediction of whether an audio artifact is likely to occur. The process can use a second learning algorithm, which also can be a DNN, to generate recommended system adjustments that can attempt to prevent the audio glitch from occurring. The recommendations can be for various systems and components on the device, such as changing the processing system frequency, the memory frequency, and the audio buffer size. After the audio artifact has been prevented, the system adjustments can be reversed fully or in steps to return the system to its state prior to the system adjustments.
1. A method of processing computer generated audio, comprising:
determining, using a neural network, a probability of an audio artifact during output of an audio signal generated using a computing system, wherein the determining the probability of the audio artifact is based at least on a processing frequency of a processor or processing unit of the system; and
adjusting at least one of the processing frequency or a memory frequency of the system to lower the probability of an audio artifact occurring when the probability is above a threshold.
2. The method of claim 1 , wherein the system is a part of a mobile device, and the audio signal is a streamed audio signal.
3. The method of claim 1 , wherein the audio signal includes at least one synthetically generated spoken phrase.
4. The method of claim 1 , wherein the adjusting includes increasing the processing frequency until the processing frequency reaches a maximum processing frequency value.
5. The method of claim 1 , wherein the adjusting includes increasing the memory frequency until the memory frequency reaches a maximum memory frequency value.
6. The method of claim 1 , further comprising reversing a prior adjustment of the system when the probability of an audio artifact occurring is not higher than the threshold.
7. The method of claim 1 , wherein the neural network is a deep neural network.
8. The method of claim 1 , wherein the adjusting includes increasing the processing frequency and the method further includes adjusting a size of an audio buffer of the system in response to increasing the processing frequency.
9. A method of processing computer generated audio, comprising:
determining, using a neural network, a probability of an audio artifact during output of an audio signal generated using a computing system, wherein the determining the probability of the audio artifact is based at least on a memory frequency of the system; and
adjusting at least one of the memory frequency of the system or a processing frequency of the system to lower the probability of an audio artifact occurring when the probability is above a threshold.
10. The method of claim 9 , wherein the system is a part of a mobile device, and the audio signal is a streamed audio signal.
11. The method of claim 9 , wherein the audio signal includes at least one synthetically generated spoken phrase.
12. The method of claim 9 , wherein the adjusting includes increasing the processing frequency until the processing frequency reaches a maximum processing frequency value.
13. The method of claim 9 , wherein the adjusting includes increasing the memory frequency until the memory frequency reaches a maximum memory frequency value.
14. The method of claim 9 , further comprising reversing a prior adjustment of the system when the probability of an audio artifact occurring is not higher than the threshold.
15. The method of claim 9 , wherein the neural network is a deep neural network.
16. A computing system comprising:
one or more processing units to perform operations that include:
determining, using a neural network, a probability of an audio artifact during output of a computer generated audio signal, wherein the determining the probability of the audio artifact is based at least on a memory frequency of the computing system; and
adjusting at least one of a processing frequency of the computing system or the memory frequency when the probability is higher than a threshold.
17. The computing system of claim 16 , wherein the neural network is a deep neural network.
18. The computing system of claim 16 , wherein the computing system is part of a mobile device, and the audio signal is a streamed audio signal.
19. The computing system of claim 16 , wherein the computing system is a vehicle computing system.
20. The computing system of claim 16 , wherein the audio signal includes one or more synthetically generated spoken phrases.