Multi-speaker conferencing over narrowband channels

• US 5,272,698 A
• Filed: 07/02/1992
• Issued: 12/21/1993
• Est. Priority Date: 09/12/1991
• Status: Expired due to Fees

• Current Assignee: United States Of America As Represented By The Secretary Of The Air Force
• Original Assignee: United States Of America As Represented By The Secretary Of The Air Force
• Inventors: Champion, Terrence G.
• Primary Examiner(s): Olms, Douglas W.
• Assistant Examiner(s): BLUM, RUSSELL
• Application Number:
  US07/909,503
• Time in Patent Office: 537 Days
• Field of Search: 341/61, 379/158, 379/202, 379/206, 370/62, 370/79, 370/84, 370/85.13, 375/122, 381/29-31
• US Class Current: 370/260
• CPC Class Codes: H04M 3/561 by multiplexing

Abstract

A system for digital conferencing over narrowband channels which allows use bymultiple simultaneous speakers and which allow for interfacing between vocoders which operate at different bit rates. The system takes advantage of the properties of multi-rate parametric vocoders (which includes the Sinusoidal Transform Coder and the Multi-Band Excitation Vocoder, as well as embedded coders), defined as parametric vocoders for whom the basic model and parameter set remain unchanged over a wide range of bit-rates. The system performs signal summation. To maintain quality for a single speaker while allowing multiple speakers, the system adaptively allocates channel bandwidth based on the number of speakers to be represented. This system also provides digital-to-digital conversion between narrow-band digitizers (vocoders) operating at different bit-rates. The system takes advantage of the characteristics of a particular class of coders, sine-wave based coders, which perform parameter estimation by sine-wave summation. Digital-to-digital conversion is performed by estimating the parameter set associated with the lower bit-rate digitizer from the parameter set associated with the higher bit-rate.

Claims

1. A system for facilitating conferencing between users of a plurality of digitally encoded voice systems which each output encoded digital signals at their own data stream bit rates, wherein said system comprises:

an electrical network which is electrically connected with each of said digitally encoded voice systems and conduct said encoded digital signals therebetween; and

a means for digital-to-digital conversion which is connected to said electrical network to receive and adjust the data stream bit rate of said encoded digital signals to enable each of the digitally encoded voice systems receive the encoded digital signals at its data stream bit rate, wherein said conversion means comprises a computer which receives digitized acoustic waveforms with coded values representing acoustic parameters that can include;

pitch, synchronization, voicing and a spectral envelope, and wherein said computer is programmed to reencode the spectral envelope with cepstral coefficients that have and adjusted bit-rate and yield thereby the encoded digital signals at the data stream bit rates of each of the digitally encoded voice systems.

2. A process for facilitating conferencing between users of a plurality of digitally encoded voice systems which each output encoded digital signals at their own data stream bit rates, wherein said process comprises:

measuring each of the particular data stream bit rates produced by each of the digitally encoded voice systems to produce thereby a set of data stream bit rates which are identified for each digitally encoded voice system; and

combining and converting all received encoded digital signals into a set of converted encoded digital signals which are each sent back to said digitally encoded voice systems at their data stream bit rate,

wherein said combining and converting step is performed by a computer which receives digitized acoustic waveforms with coded values representing parameters that can include;

pitch synchronization, voicing and a spectral envelope, and wherein said computer is programmed to perform parameter de-scoping and change spacing between the coded values so that each of said digitally encoded voice systems receives the digitized acoustic waveforms at its data stream bit rate.

3. A process, as defined in claim 2, wherein said digitally encoded voice systems are each narrow-band vocoders which each include digitized acoustic voice data in their encoded digital signals by performing sinusoidal transform coding and wherein said converting step is performed by a computer which receives digitized acoustic waveforms with coded values representing acoustic parameters that can include:

pitch, synchronization, voicing and spectral envelope, and wherein said computer is programmed to reencode the spectral envelope with cepstral coefficients that have an adjusted bit-rate and yield thereby the encoded digital signals at the data stream bit rates of each of the digitally encoded voice systems.