Telephone network apparatus and method using echo delay and attenuation

• US 5,761,638 A
• Filed: 03/17/1995
• Issued: 06/02/1998
• Est. Priority Date: 03/17/1995
• Status: Expired due to Term

• Current Assignee: Qwest Communications International Incorporated (Lumen Technologies, Inc.)
• Original Assignee: MediaOne Group Inc. (Comcast Corporation)
• Inventors: Jaramillo, Paul D., Knittle, Curtis D., Wu, Frank H.
• Primary Examiner(s): Hudspeth, David R.
• Assistant Examiner(s): SAX, ROBERT L
• Application Number:
  US08/406,352
• Time in Patent Office: 1,173 Days
• Field of Search: 370/32.1, 379/88, 379/67, 379/410, 379/411, 704/233
• US Class Current: 704/233
• CPC Class Codes: G10L 2021/02082 the noise being echo, reverberation of the speech
• CPC Class Codes: H04B 3/23 using a replica of transmitted signal in the time domain, e.g. echo cancellers
• CPC Class Codes: H04M 2201/40 using speech recognition speech recognition per se G10L15/00
• CPC Class Codes: H04M 3/002 Applications of echo suppressors or cancellers in telephonic connections in two-way loud-speaking telephone systems H04M9/08, echo suppressors or cancellers per se H04B3/20
• CPC Class Codes: H04M 3/493 Interactive information services, e.g. directory enquiries ; Arrangements therefor, e.g. interactive voice response [IVR] systems or voice portals

Abstract

The present invention provides a telephone network apparatus for performing speech recognition services in a telephone system in substantially real time. The apparatus uses a telephone channel signal to determine the echo delay of a telephone channel and then uses this delay to configure an echo cancellation filter for use in performing speech recognition. Use of echo delay in configuring the filter allows the echo cancellation function to be done using much less computational time than would be needed without its use, thereby granting a speech recognition unit greater access to a resident microprocessor to perform its function in substantially real time.

Claims

1. A method for providing speech recognition in a telephone system, comprising:

receiving a telephone signal from a first of a plurality of user communication devices;

sending a chirp signal to said first user communication device;

receiving an echo signal based on said chirp signal, obtaining attenuation data using at least one of said chirp signal and said echo signal;

determining whether echo cancellation should be performed, using said attenuation data;

obtaining a time delay using said chirp signal and said echo signal;

determining a value for a bulk delay filter element using said time delay;

using an adaptive filter to determine one or more filter coefficient values for performing echo cancellation;

configuring echo cancellation means using at least said value for said bulk delay filter element and said one or more filter coefficient values;

receiving a speech signal from said first user communication device after said configuring step; and

recognizing said speech signal;

wherein said steps of obtaining attenuation data and determining whether echo cancellation should be performed are performed after said step of receiving said echo signal.

2. A method, as claimed in claim 1, wherein:

said step of obtaining said attenuation data includes;

matched filtering said echo signal with a time reversed version of said chirp signal to create an output signal; and

obtaining an amplitude of the largest peak of said output signal and an amplitude of the next largest peak of said output signal; and

said step of determining whether echo cancellation should be performed includes;

determining a ratio using said amplitude of the largest peak and said amplitude of the next largest peak; and

comparing said ratio to a threshold value.

3. A method, as claimed in claim 1, wherein:

said step of obtaining said attenuation data includes;

matched filtering said echo signal with a time reversed version of said chirp signal to create an output signal; and

obtaining an amplitude of the largest peak of said output signal and an amplitude of said chirp signal; and

said step of determining whether echo cancellation should be performed includes;

determining a ratio using said amplitude of the largest peak of said output signal and said amplitude of said chirp signal; and

comparing said ratio to a threshold value.

4. An apparatus, as claimed in claim 3, wherein:

said step of determining a ratio includes;

determining an estimate of an amplitude of said echo signal using said amplitude of the largest peak of said output signal and said amplitude of said chirp signal; and

determining a ratio of said estimate of said amplitude of said echo signal to said amplitude of said chirp signal.

5. An apparatus for providing speech recognition in a telephone system comprising:

a plurality of user communication devices including a first user communication device;

a central unit responsive to said plurality of user communication devices through which speech and non-speech signals are communicated;

a telephone network apparatus communicating with said central unit said telephone network apparatus including;

first means for sending a telephone channel signal for evaluating a telephone channel communicating with said first user communication device;

second means for obtaining a magnitude related to a time delay based on said telephone channel signal and an echo signal related to said telephone channel signal;

an adaptive filter, responsive to said second means for determining values for a bulk delay parameter and one or more variable tap coefficients;

an echo cancellation unit electrically connected to said adaptive filter for canceling an echo signal received from said central unit;

means for obtaining attenuation data using at least one of said telephone channel signal and said echo signal;

means for determining whether echo cancellation should be performed based on said attenuation data;

speech recognition means for recognizing a speech signal from said first user communication device; and

control means for controlling sending and receiving of signals relative to said central unit and said user devices.

6. An apparatus, as claimed in claim 5, said telephone network apparatus further including:

means for bypassing said adaptive filter and said echo cancellation unit whenever said means for determining determines that echo cancellation should not be performed.

7. An apparatus, as claimed in claim 5, wherein:

said means for obtaining attenuation data includes;

a matched filter for filtering said echo signal with a time reversed version of said telephone channel signal to create an output signal; and

means for obtaining an amplitude of the largest peak of said output signal and an amplitude of the next largest peak of said output signal; and

said means for determining includes;

means for determining a ratio using said amplitude of the largest peak and said amplitude of the next largest peak; and

means for comparing said ratio to a threshold value.

8. An apparatus, as claimed in claim 5, wherein:

said means for obtaining attenuation data includes;

a matched filter for filtering said echo signal with a time reversed version of said telephone channel signal to create an output signal; and

means for obtaining an amplitude of the largest peak of said output signal and an amplitude of said telephone channel signal; and

said means for determining includes;

means for determining a ratio using said amplitude of the largest peak of said output signal and said amplitude of said telephone channel signal; and

means for comparing said ratio to a threshold value.

9. An apparatus, as claimed in claim 8, wherein:

said means for determining a ratio includes;

means for determining an estimate of an amplitude of said echo signal using said amplitude of the largest peak of said output signal and said amplitude of said telephone channel signal; and

means for determining a ratio of said estimate of said amplitude of said echo signal to said amplitude of said telephone channel signal.