Back to results
US 12,711,974 B2Granted
Method, system, and computer program product for dynamically improving voice quality in mobile cellular communication network
AssigneeP.I. Works U.S., Inc.
InventorsAhmet Serkan Sofuoglu, Yakup Tarik Kranda, Ubeyde Mavuş, Sina Alp
Life of this patent
Estimating the term…
Abstract
Voice data packets are analyzed to determine the audio codec mode and one or more data metrics associated with the voice data packets in a cellular voice call that includes a cellular mobile device in cellular communication with a base station. The audio codec mode and the data metric(s) is/are provided to a trained machine-learning model to determine a predicted mean-opinion score for the voice call. One or more network parameters are updated on the base station for the cellular user equipment device to improve the data metric(s) and the perceived quality of the cellular voice call.
Claims14
A computer-implemented method for dynamically improving a quality of a cellular voice call, comprising:
- receiving voice packet data associated with the cellular voice call that includes a cellular mobile device in cellular communication with a base station, the voice packet data including a plurality of voice packets and representing a measurement time window, the measurement time window subdivided into a plurality of sub-measurement windows;
- determining an audio codec mode of the voice packet data and a number of silent indicator descriptor (SID) packets in the voice packet data based, at least in part, on an average packet length of voice packets in each sub-measurement window, a known length of SID frames in the voice packet data, and a set of predetermined packet lengths associated with a speech frame, each predetermined packet length corresponding to a respective known voice codec mode, each SID packet including at least one SID frame;
- determining one or more real-time data metrics of SID-free packets in at least some of the sub-measurement windows, the SID-free packets only including speech frames with the number of SID packets equal to 0;
- determining a predicted mean-opinion score (MOS) of the cellular voice call by feeding the real-time data metric(s) and the audio codec mode into a trained machine-learning model, the trained machine-learning model having been trained with known MOS values and known data metrics of test SID-free packets from test calls and known audio codec modes;
- and
- dynamically modifying a base-station configuration for the cellular mobile device to improve at least one of the real-time data metric(s) so as to improve an end-user perceived speech quality of the cellular voice call associated with the audio codec mode.
3 more elements
A speech detriment management system (SDMS) comprising a non-transitory computer-readable medium comprising instructions that, when executed by a processor, cause the processor to:
- receive voice packet data associated with the cellular voice call that includes a cellular mobile device in cellular communication with a base station, the voice packet data including a plurality of voice packets and representing a measurement time window, the measurement time window subdivided into a plurality of sub-measurement windows;
- determine an audio codec mode of the voice packet data and a number of silent indicator descriptor (SID) packets in the voice packet data based, at least in part, on an average packet length of voice packets in each sub-measurement window, a known length of SID frames in the voice packet data, and a set of predetermined packet lengths associated with a speech frame, each predetermined packet length corresponding to a respective known voice codec mode, each SID packet including at least one SID frame;
- determine one or more real-time data metrics of SID-free packets in at least some of the sub-measurement windows, the SID-free packets only including speech frames with the number of SID packets equal to 0;
- determine a predicted mean-opinion score (MOS) of the cellular voice call by feeding the real-time data metric(s) and the audio codec mode into a trained machine-learning model, the trained machine-learning model having been trained with known MOS values and known data metrics of test SID-free packets from test calls and known audio codec modes;
- and
- dynamically modify a base-station configuration for the cellular mobile device to improve at least one of the real-time data metric(s) so as to improve an end-user perceived speech quality of the cellular voice call associated with the audio codec mode.
3 more elements
A non-transitory computer-readable medium comprising instructions that, when executed by a processor, cause the processor to:
- receive voice packet data associated with the cellular voice call that includes a cellular mobile device in cellular communication with a base station, the voice packet data including a plurality of voice packets and representing a measurement time window, the measurement time window subdivided into a plurality of sub-measurement windows;
- determine an audio codec mode of the voice packet data and a number of silent indicator descriptor (SID) packets in the voice packet data based, at least in part, on an average packet length of voice packets in each sub-measurement window, a known length of SID frames in the voice packet data, and a set of predetermined packet lengths associated with a speech frame, each predetermined packet length corresponding to a respective known voice codec mode, each SID packet including at least one SID frame;
- determine one or more real-time data metrics of SID-free packets in at least some of the sub-measurement windows, the SID-free packets only including speech frames with the number of SID packets equal to 0;
- determine a predicted mean-opinion score (MOS) of the cellular voice call by feeding the real-time data metric(s) and the audio codec mode into a trained machine-learning model, the trained machine-learning model having been trained with known MOS values and known data metrics of test SID-free packets from test calls and known audio codec modes;
- and
- dynamically modify a base-station configuration for the cellular mobile device to improve at least one of the real-time data metric(s) so as to improve an end-user perceived speech quality of the cellular voice call associated with the audio codec mode.
3 more elements
Dependencies are inferred from the claim text.