METHOD AND APPARATUS OF VOICE MIXING FOR CONFERENCING AMONGST DIVERSE NETWORKS
First Claim
Patent Images
1. An apparatus for performing voice mixing of multiple inputs from multiple source bit-streams representing frames of data from a plurality of source channels, each of the plurality of source channels being connected to a conference and encoded according to a codec employed by each of the plurality of source channels, the apparatus comprising:
- a bit-stream un-packer for each of the plurality of source channels, each of the plurality of source channels being connected to a mixing system;
a voice activity detection module for each of the plurality of source channels, wherein the voice activity detection module is adapted to determine if an input channel is active;
a decision module adapted to determine if an output on a first channel of the plurality of source channels connected to the conference should be obtained through time domain mixing of time domain signals associated with other channels of the plurality of source channels or through fast transcoding of one of the other channels of the plurality of source channels;
a switch module adapted to connect an input from one of the plurality of source channels to at least one of an interpolator module or a time domain mixing module based on the determined output;
an interpolator module between each of the plurality of source channels and adapted to allow speech compression parameters produced by one speech compression algorithm to cover a given time period and to represent a time period that another speech compression algorithm utilizes;
a time domain mixing module for each of the plurality of source channels, wherein the time domain mixing module is adapted to produce a time domain signal that represents a combination of the time domain signals associated with other channels of the plurality of source channels; and
a pack module for each of the plurality of source channels, wherein the pack module is adapted to provide a resultant conference signal in a format associated with an output of at least one of the plurality of source channels.
5 Assignments
0 Petitions
Accused Products
Abstract
A conferencing system is provided that utilizes both time domain signal mixing and direct signal fast transcoding. An exemplary embodiment of the present invention utilizes both time domain signal mixing and direct signal fast transcoding to process a bit-stream from a same channel during a conference.
37 Citations
39 Claims
-
1. An apparatus for performing voice mixing of multiple inputs from multiple source bit-streams representing frames of data from a plurality of source channels, each of the plurality of source channels being connected to a conference and encoded according to a codec employed by each of the plurality of source channels, the apparatus comprising:
-
a bit-stream un-packer for each of the plurality of source channels, each of the plurality of source channels being connected to a mixing system;
a voice activity detection module for each of the plurality of source channels, wherein the voice activity detection module is adapted to determine if an input channel is active;
a decision module adapted to determine if an output on a first channel of the plurality of source channels connected to the conference should be obtained through time domain mixing of time domain signals associated with other channels of the plurality of source channels or through fast transcoding of one of the other channels of the plurality of source channels;
a switch module adapted to connect an input from one of the plurality of source channels to at least one of an interpolator module or a time domain mixing module based on the determined output;
an interpolator module between each of the plurality of source channels and adapted to allow speech compression parameters produced by one speech compression algorithm to cover a given time period and to represent a time period that another speech compression algorithm utilizes;
a time domain mixing module for each of the plurality of source channels, wherein the time domain mixing module is adapted to produce a time domain signal that represents a combination of the time domain signals associated with other channels of the plurality of source channels; and
a pack module for each of the plurality of source channels, wherein the pack module is adapted to provide a resultant conference signal in a format associated with an output of at least one of the plurality of source channels. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22)
-
-
23. A method for performing voice mixing of multiple inputs from multiple source bit-streams representing frames of data from a plurality of source channels, each of the plurality of source channels being connected to a conference and encoded according to a codec employed by each of the plurality of source channels, the method comprising:
-
un-packing input compression codes from the multiple source bit-streams, wherein the multiple source bit-streams represent encoded signals;
detecting a voice activity present on each of the plurality of source channels for a pre-set time period in an adaptable manner;
reconstructing time domain signals from voice active input source bit-streams that are from source channels other than a first output channel of the plurality of source channels;
mixing the reconstructed time domain signals into a mixed output signal;
generating compression codes representing the mixed output signal;
interpolating input compression codes from a single voice active bit-stream from a first source channel to output compression codes to be placed on a second channel of the plurality of source channels connected to the conference when only a single source channel, other than the second, is detected to have voice activity; and
packing the output compression codes in an output bit-stream formatted to represent frames of data to be placed on a channel of the plurality of source channels. - View Dependent Claims (24, 25, 26, 27, 28, 29, 30, 31, 32, 33)
-
- 34. A conferencing system adapted to conference a number of channels such that no restrictions are placed on the type of compression used by any of the channels in that the system includes modules that can unpack bit-streams of numerous compression standards.
- 37. A conferencing system that utilizes both time domain signal mixing and direct signal fast transcoding.
-
39. A conferencing system that allows a session which performs transcoding in code space to become a conferencing session and vice versa without the need for the conferencing and transcoding functionalities to be split between different systems.
Specification