Patent US5050160A successfully refreshed from Google Patents.
← Back to Patent Explorer
US5050160A Expired / Lapsed US ⏱️ 0.0 years remaining

Training method for an echo canceller for use in a voice conference system

Assignee: NEC Corp
•
Filing: 1990-06-12
•
Granted: 1991-09-17
•
Expires: 2010-06-12
📄 Original PDF 🌐 Google Patents 🔄 Sync Now

Abstract

Abstract A method for training an echo canceller of a voice conference system, to cancel an echo by subtracting an estimated echo signal from an actual echo signal, includes the steps of: generating, at a time of a first training session, an initial echo signal and an initial training signal through feeding of a training signal to a loudspeaker and to the echo canceller; obtaining a response characteristic of the echo canceller to the echo signals; storing the initial echo signal and the initial training signal; and training the echo canceller, during subsequent training sessions, by using the stored initial echo signal and the stored training signal.

Inventors

  • 👤 Hitoshi Fuda

Bibliographic Details

Application Number —
Jurisdiction US
Data Source google_patents
Last Synchronized 2026-11-02 12:07
CPC Classifications
H; H04; H04M; H04M9/00; H04M9/08; H04M9/082

Claims Analysis

Total Claims: 4
Claim 1
Independent Claim 131 words
A method of training an echo canceller for cancelling an echo by subtracting an estimated echo signal generated in response to a receive-in signal fed to a loudspeaker from an echo signal which is generated by a microphone in response to said receive-in signal, said method comprising the steps of: generating, at the time of a first training, an initial echo signal by feeding an initial training signal to the loudspeaker; training the echo canceller in response to the initial echo signal and the initial training signal to obtain a response characteristic of the echo canceller to the echo signal; storing the initial echo signal and the initial training signal; and training, after the first training, the echo canceller by using the stored initial echo signal and the stored training signal.
Claim 2
Independent Claim 131 words
A method of training an echo canceller for cancelling an echo by subtracting an estimated echo signal generated in response to a receive-in signal fed to a loudspeaker from an echo signal which is generated by a microphone in response to said receive-in signal, said method comprising the steps of: generating, at the time of a first training, an initial echo signal by feeding an initial training signal to the loudspeaker; training the echo canceller in response to the initial echo signal and the initial training signal to obtain a response characteristic of the echo canceller to the echo signal; storing the initial echo signal and the initial training signal; and training, after the first training, the echo canceller by using the stored initial echo signal and the stored training signal.
Claim 3
Dependent Claim 177 words
A voice conference system comprising: send-out signal generating means having microphone means for producing a send-out signal; receive-in signal reproducing means for causing a loudspeaker to reproduce a receive-in signal coming in over a communication line to thereby output a reproduced signal; echo canceller means for cancelling, in response to the receive-in signal, an echo signal ascribable to the microphone means which picks up the reproduced signal; training signal generating means for generating a training signal for said echo canceller means; first storing means for storing the training signal; second storing means for storing a training echo signal which the microphone means produces by picking up a reproduced training signal which the loudspeaker has generated by reproducing the training signal; and selector means for selecting, when said echo canceller means is to be trained for the first time, the training echo signal and the training signal and, when said echo canceller means is to be trained afterwards, selecting a training signal and a training echo signal which are stored in said first and second storing means, respectively.
Claim 4
Dependent Claim 177 words
A voice conference system comprising: send-out signal generating means having microphone means for producing a send-out signal; receive-in signal reproducing means for causing a loudspeaker to reproduce a receive-in signal coming in over a communication line to thereby output a reproduced signal; echo canceller means for cancelling, in response to the receive-in signal, an echo signal ascribable to the microphone means which picks up the reproduced signal; training signal generating means for generating a training signal for said echo canceller means; first storing means for storing the training signal; second storing means for storing a training echo signal which the microphone means produces by picking up a reproduced training signal which the loudspeaker has generated by reproducing the training signal; and selector means for selecting, when said echo canceller means is to be trained for the first time, the training echo signal and the training signal and, when said echo canceller means is to be trained afterwards, selecting a training signal and a training echo signal which are stored in said first and second storing means, respectively.

References Cited (Backward Citations)

US4621173A Cited
US4852161A Cited
US4827473A Cited
US4891801A Cited
US4896318A Cited

USPTO Annuity & Maintenance Windows

Annuity Stage Window Opens Payment Deadline Surcharge Period Until Estimated Fee Status
3.5 Years (Year 4) 1994-09-16 1995-03-18 1995-09-17 $2000.00 Window Closed / Lapsed
7.5 Years (Year 8) 1998-09-16 1999-03-18 1999-09-17 $3760.00 Window Closed / Lapsed
11.5 Years (Year 12) 2002-09-16 2003-03-18 2003-09-17 $7700.00 Window Closed / Lapsed

Assigned Categories & Tags

No categories assigned to this patent yet.

Assign Category Tag