Complexity resource manager for multi-channel speech processing
View Patent ↗A multi-channel speech processor for encoding speech in a packet network environment is disclosed. In one illustrative aspect, a complexity resource manager (CRM) is executed by a controller or processor. The CRM manages the level of complexity of encoding which is used by a signal processing unit (SPU) to convert the speech signal into packet data. In general, the CRM determines the level of complexity of encoding based on a calculated complexity budget, where the complexity budget is determined based on the time required to process prior speech signal channels and the time available to process the remaining channels. In this way, the CRM is able to control the overall complexity of the speech processor through its ability to signal the SPU to encode speech signal in a complexity reduced mode based on the calculated complexity budget under certain conditions.
1. A method for managing a processing time of a speech processor having at least one signal processing unit (SPU) configured to encode speech signals, the speech processor capable of interfacing with a plurality of communication channels, the speech processor having a time budget during which to process speech signals from each of the plurality of communication channels at least once, the method comprising:
receiving an input speech signal from one of the communication channels;
determining a total amount of time previously spent for processing one or more of the communication channels before the receiving the input speech signal from the one of the communication channels;
calculating a remaining time budget based on the total amount of time previously spent and the time budget; and
signaling the SPU to change a present coding rate to a new coding rate based on the remaining time budget and a remaining number of communication channels of the plurality of communication channels to be processed;
wherein the SPU selects the new coding rate such that the speech processor is able to process speech signals from the remaining number of communication channels of the plurality of communication channels at least once during the remaining time budget.
2. The method of claim 1 , wherein the SPU is further configured to encode the input speech signal using the new coding rate selected from a plurality of coding rates of a speech coder, and wherein at least two of the plurality of coding rates have different levels of complexity.
3. The method of claim 1 , wherein the new coding rate is of lower complexity than the present coding rate.
4. A method for managing a processing time of a speech processor having at least one signal processing unit (SPU) configured to encode speech signals, the speech processor capable of interfacing with a variable number of a plurality of communication channels, the speech processor having a time budget during which to process speech signals from each of the plurality of communication channels at least once, the method comprising:
receiving a request to process speech signals from a new communication channel;
determining a total amount of time previously spent for processing one or more of the communication channels before the receiving the request to process speech signals from the new communication channel;
calculating a remaining time budget based on the total amount of time previously spent and the time budget; and
accepting the request if the speech processor determines that the speech processor is able to process the new communication channel in addition to a remaining number of communication channels at least once during the remaining time budget.
5. The method of claim 4 further comprising: denying the request if the speech processor determines that the speech processor is not able to process the new communication channel in addition to the remaining number of communication channels at least once during the remaining time budget.
6. The method of claim 4 further comprising:
signaling the SPU to change a present coding rate to a new coding rate based on the remaining time budget, and new communication channel and the remaining number of communication channels of the plurality of communication channels to be processed;
wherein the SPU selects the new coding rate such that the speech processor is able to process speech signals from the new communication channel in addition to the remaining number of communication channels of the plurality of communication channels at least once during the remaining time budget.
7. A multi-channel speech processor configured to process speech signals from a plurality of communication channels within a time budget, the multi-channel speech processor comprising:
a controller capable of interfacing with the plurality of communication channels, wherein the controller is configured to receive an input speech signal from one of the communication channels, and further configured to determine a total amount of process time previously spent for processing one or more of the communication channels before receiving the input speech signal from the one of the communication channels;
a memory coupled to the controller configured to store process time values, wherein each of the process time values is indicative of an amount of time spent for processing input speech signals from one or more of the communication channels; and
at least one signal processing unit (SPU) coupled to the memory and to the controller, the SPU configured to encode the speech signals;
wherein the controller calculates a remaining time budget based on the total amount of time previously spent and the time budget and wherein the controller signals the SPU to change a present coding rate to a new coding rate based on the remaining time budget and a remaining number of communication channels of the plurality of communication channels to be processed, and wherein the SPU selects the new coding rate such that the speech processor is able to process speech signals from the remaining number of communication channels of the plurality of communication channels at least once during the remaining time budget.
8. The multi-channel speech processor of claim 7 , wherein the SPU is further configured to encode the input speech signal using the new coding rate selected from a plurality of coding rates of a speech coder, and wherein at least two of the plurality of coding rates have different levels of complexity.
9. The multi-channel speech processor of claim 7 , wherein the new coding rate is of lower complexity than the present coding rate.
10. A multi-channel speech processor configured to process speech signals from a plurality of communication channels within a time budget, the multi-channel speech processor comprising:
a controller capable of interfacing with the plurality of communication channels, wherein the controller is configured to receive a request to process speech signals from a new communication channel, and further configured to determine a total amount of time previously spent for processing one or more of the communication channels before receiving the request to process speech signals from the new communication channel;
a memory coupled to the controller configured to store process time values, wherein each of the process time values is indicative of an amount of time spent for processing input speech signals from one or more of the communication channels; and
at least one signal processing unit (SPU) coupled to the memory and to the controller, the SPU configured to encode the speech signals;
wherein the controller is configured to calculate a remaining time budget based on the total amount of time previously spent and the time budget, and wherein the controller accepts the request if the controller determines that the speech processor is able to process the new communication channel in addition to a remaining number of communication channels at least once during the remaining time budget.
11. The multi-channel speech processor of claim 10 wherein the controller denies the request if the controller determines that the speech processor is not able to process the new communication channel in addition to the remaining number of communication channels at least once during the remaining time budget.
12. The multi-channel speech processor of claim 10 , wherein the controller signals the SPU to change a present coding rate to a new coding rate based on the remaining time budget, and new communication channel and the remaining number of communication channels of the plurality of communication channels to be processed, and wherein the SPU selects the new coding rate such that the speech processor is able to process speech signals from the new communication channel in addition to the remaining number of communication channels of the plurality of communication channels at least once during the remaining time budget.