Method and apparatus for performing active packet bundling in a voice over IP communications system based on voice concealability
View Patent ↗A method and apparatus for performing a active packet bundling in a VoIP (Voice over Internet Protocol) based on a measure of voice concealability of one or more speech packets to be transmitted. Voice concealability is a property of a speech packet which indicates the likelihood that a given speech segment will meet the “quasi-stationary” (QS) behavior assumed by conventional Packet Loss Concealment (PLC) algorithms used to compensate for transmission loss of speech packets. By ensuring that only well-concealed frames are delayed for bundling, it is more likely that the receiver will run the PLC algorithm when it is most effective (i.e., when it is likely to perform well at reconstructing the missing speech frame), rather than when it is more likely to be ineffective. This allows for a reduction in bandwidth while minimizing any potential negative impact on quality.
1. A method for performing packet bundling of voice frames comprised in a sequence of voice frames to be transmitted across a Voice-over IP communications network, the method comprising the steps of:
determining, with use of a processor, whether or not each of a plurality of voice frames comprised in said sequence of voice frames is concealable, said determination being made, based on a quantitative analysis of speech waveform energy, for each of said plurality of voice frames independently with respect to the other voice frames of said plurality of voice frames in said sequence, wherein the quantitative analysis includes counting a number of bumps in the speech waveform energy;
assigning a score to each of said plurality of voice frames based on the quantitative analysis;
bundling, with use of the processor, a given one of said voice frames together with at least one subsequent voice frame in said sequence of voice frames when the score of the given one of said voice frames has been determined to satisfy a threshold; and
transmitting, with use of a transmitter, said bundled voice frames together across said communications network as a single packet; and
transmitting, with use of a transmitter, one or more other ones of said plurality of voice frames across said communications network without bundling said other voice frames together with subsequent voice frames, when the scores of said other voice frames have been determined not to satisfy a threshold.
2. The method of claim 1 further comprising the step of counting a number of voice frames which have been bundled together and transmitting said bundled voice frames together across said communications network as a single packet when said number of voice frames which have been bundled together satisfies a threshold.
3. The method of claim 1 wherein said step of determining whether or not the particular voice frame comprised in said sequence of voice frames is concealable is based on an associated packet loss concealment algorithm.
4. The method of claim 3 further comprising the steps of:
executing the associated packet loss concealment algorithm based on a hypothetical assumption that the particular voice frame has been lost; and
calculating said measure of voice concealability based on said execution of the associated packet loss concealment algorithm.
5. The method of claim 1 wherein said step of determining whether or not a particular voice frame comprised in said sequence of voice frames is concealable comprises calculating a measure of voice concealability based on a quasi-stationary failure estimate which comprises a phoneme rate estimate.
6. The method of claim 5 wherein the phoneme rate estimate is determined based on the quantitative analysis of speech waveform energy.
7. An apparatus for performing packet bundling of voice frames comprised in a sequence of voice frames to be transmitted across a Voice-over IP communications network, the apparatus comprising:
a processor which determines whether or not each of a plurality of voice frames comprised in said sequence of voice frames is concealable, said determination being made, based on a quantitative analysis of speech waveform energy, for each of said plurality of voice frames independently with respect to the other voice frames of said plurality of voice frames in said sequence, wherein the quantitative analysis includes counting a number of bumps in the speech waveform energy, which assigns a score to each of said plurality of voice frames based on the quantitative analysis, and which bundles a given one of said voice frames together with at least one subsequent voice frame in said sequence of voice frames when the score of the given one of said voice frames has been determined to satisfy a threshold; and
a transmitter which transmits said bundled voice frames together across said communications network as a single packet, and which transmits one or more other ones of said plurality of voice frames across said communications network without bundling said other voice frames together with subsequent voice frames, when the scores of said other voice frames have been determined not to satisfy a threshold.
8. The apparatus of claim 7 wherein said processor further counts a number of voice frames which have been bundled together, and wherein said transmitter further transmits said bundled voice frames together across said communications network as a single packet when said number of voice frames which have been bundled together satisfies a threshold.
9. The apparatus of claim 7 wherein said processor determines whether or not the particular voice frame comprised in said sequence of voice frames is concealable based on an associated packet loss concealment algorithm.
10. The apparatus of claim 9 wherein said processor executes the associated packet loss concealment algorithm based on a hypothetical assumption that the particular voice frame has been lost, and calculates said measure of voice concealability based on said execution of the associated packet loss concealment algorithm.
11. The apparatus of claim 7 wherein said processor determines whether or not each of a plurality of voice frames comprised in said sequence of voice frames is concealable by calculating a measure of voice concealability based on a quasi-stationary failure estimate which comprises a phoneme rate estimate.
12. The apparatus of claim 11 wherein the phoneme rate estimate is determined based on the quantitative analysis of speech waveform energy.
13. A non-transitory computer readable medium comprising executable program code for performing packet bundling of voice frames comprised in a sequence of voice frames to be transmitted across a Voice-over IP communications network, the executable program code configured to:
determine whether or not each of a plurality of voice frames comprised in said sequence of voice frames is concealable, said determination being made, based on a quantitative analysis of speech waveform energy, for each of said plurality of voice frames independently with respect to the other voice frames of said plurality of voice frames in said sequence, wherein the quantitative analysis includes counting a number of bumps in the speech waveform energy;
assign a score to each of said plurality of voice frames based on the quantitative analysis; and
bundle a given one of said voice frames together with at least one subsequent voice frame in said sequence of voice frames when the score of the given one of said voice frames has been determined to satisfy a threshold,
wherein said bundled voice frames are configured to be transmitted together across said communications network as a single packet and wherein one or more other ones of said plurality of voice frames are configured to be transmitted across said communications network without bundling said other voice frames together with subsequent voice frames, when the scores of said other voice frames have been determined not to satisfy a threshold.
14. The computer readable medium of claim 13 wherein said executable program code is further configured to:
count a number of voice frames which have been bundled together; and
determine when said number of voice frames which have been bundled together satisfies a threshold,
wherein said bundled voice frames are configured to be transmitted together across said communications network as a single packet when said number of voice frames which have been bundled together satisfy the threshold.
15. The computer readable medium of claim 13 wherein said executable program code is configured to determine whether or not the particular voice frame comprised in said sequence of voice frames is concealable based on an associated packet loss concealment algorithm.
16. The computer readable medium of claim 15 wherein said executable program code is further configured to:
execute the associated packet loss concealment algorithm based on a hypothetical assumption that the particular voice frame has been lost, and
calculate said measure of voice concealability based on said execution of the associated packet loss concealment algorithm.
17. The computer readable medium of claim 13 wherein the executable program code is further configured to determine whether or not each of a plurality of voice frames comprised in said sequence of voice frames is concealable by calculating a measure of voice concealability based on a quasi-stationary failure estimate which comprises a phoneme rate estimate.
18. The computer readable medium of claim 17 wherein the phoneme rate estimate is determined based on the quantitative analysis of speech waveform energy.