Conference bridge for packetized speech-signal networks

• US 5,436,896 A
• Filed: 03/17/1994
• Issued: 07/25/1995
• Est. Priority Date: 03/17/1994
• Status: Expired due to Term

• Current Assignee: AT&T IPM Corp. (AT&T, Inc.)
• Original Assignee: AT&T Corporation (AT&T, Inc.)
• Inventors: Anderson, Thomas W., Vancil, Paul W., Tiedemann, Norman R.
• Primary Examiner(s): Olms, Douglas W.
• Assistant Examiner(s): HOM, SHICK C
• Application Number:
  US08/214,724
• Time in Patent Office: 495 Days
• Field of Search: 370/62, 370/94.1, 370/32.1, 370/85.13, 379/201, 379/202, 379/204, 379/158, 379/159, 379/406, 379/390
• US Class Current: 370/260
• CPC Class Codes: H04L 12/6418 Hybrid transport
• CPC Class Codes: H04L 2012/5642 Multicast/broadcast/point-multipoint, e.g. VOD
• CPC Class Codes: H04L 2012/5643 Concast/multipoint-to-point
• CPC Class Codes: H04L 2012/5671 Support of voice
• CPC Class Codes: H04L 2012/6481 Speech, voice
• CPC Class Codes: H04M 3/56 Arrangements for connecting several subscribers to a common circuit, i.e. affording conference facilities video conference systems H04N7/15
• CPC Class Codes: H04M 3/569 using the instant speaker's algorithm speech detection per se G10L25/78
• CPC Class Codes: H04Q 11/0478 Provisions for broadband connections

Abstract

A conference bridge that receives speech data in the form of data packets, and transmits data in the same form, without transforming the data in the conference bridge. The conference bridge according to this invention includes a plurality of inputs that have speech detectors that detect the presence of speech data. The speech detectors report the presence of speech to a controller. The controller causes data packets from one of the inputs detecting speech to be replicated for all outputs. If there is speech at more than one input at a time, then a decision is made as to which input to replicate. Advantageously, the decision is based on who is the loudest speaker. Further, the data that is replicated is not sent to the output for the originator in order to prevent echo.

Claims

1. A conference bridge apparatus for a packet network, said conference bridge receiving input communications in the form of data packets from a plurality of participants and distributing said data packets to said plurality of participants, said apparatus comprising:

a plurality of inputs, one of said plurality of inputs being connected to each of said plurality of participants, each of said plurality of inputs having a speech detection means for generating indications of the presence of data packets representing speech at its input;

a plurality of outputs, one of said plurality of outputs connected to each of said participants;

replicating means connected to said plurality of inputs and said plurality of outputs for receiving data packets from only one of said inputs, for replicating said received data packets; and

selector means connected to said plurality of speech detection means and said replicating means for comparing said indications from each of said speech detection means to determine which input has a speech signal and for causing said replicating means to replicate data packets from said input having a speech signal and send said replicated packets to ones of said plurality of outputs.

2. A conference bridge apparatus according to claim 1 wherein said replication means includes means for not sending packets from said input having a speech signal to an output related to said input having a speech signal.

3. A conference bridge apparatus according to claim 1 wherein said network comprises an asynchronous transfer mode network.

8. A conference bridge apparatus according to claim 1 further including an automatic gain control connected to said replicating means for controlling the gain of said data packets from said input having a speech signal before said data packets are replicated.

4. A method for use in a conference bridge in a data network, said conference bridge including a plurality of inputs connected to a plurality of participants, a plurality of speech detectors for detecting the presence of speech data at each of said plurality of inputs and generating an indication when speech is detected, a plurality of outputs connected to said plurality of participants, replicating means connected to said plurality of inputs and said plurality of outputs for receiving data packets containing said speech data from said inputs, replicating said data packets and sending them to said plurality of outputs, and selector means connected to said plurality of speech detectors and said replicating means for controlling said replicating means, said method comprising the steps of:

said selector means determining which of said plurality of inputs has a speech indication from ones of said plurality of speech detectors;

said selector means causing said replicating means to replicate data packets from said input having a speech signal; and

said replicating means sending said data packets to said plurality of outputs.

5. A method for use in a conference bridge according to claim 4 wherein said replicating means does not send said replicated data packets to an output related to said input having a speech signal.

6. A method for use in a conference bridge according to claim 4 wherein said step of determining which of said plurality of inputs has a speech indication comprises determining which of said speech indications are the loudest.

7. A method for use in a conference bridge according to claim 4 wherein said controller means causes said replicating means to replicate data packets from only one input at a time.