Voice quality performance evaluator and method of operation in conjunction with a communication network

• US 6,330,428 B1
• Filed: 12/23/1998
• Issued: 12/11/2001
• Est. Priority Date: 12/23/1998
• Status: Expired due to Fees

• Current Assignee: Northern Telecom Limited (Nortel Networks Corporation)
• Original Assignee: Nortel Networks Limited (Nortel Networks Corporation)
• Inventors: Cooney, Chris G., Vea, Matthew J. J., Lewis, Larry D.
• Primary Examiner(s): Trost, William
• Assistant Examiner(s): Gesesse, Tilahun
• Application Number:
  US09/220,219
• Time in Patent Office: 1,084 Days
• Field of Search: 455/67.1, 455/423, 455/67.2, 455/67.3, 455/67.4, 455/67.6, 455/424, 455/425, 375/224, 370/249, 370/250, 370/241, 704/230, 704/212, 704/224, 379/1.02, 379/1.03
• US Class Current: 455/67.11
• CPC Class Codes: H04B 7/18513 Transmission in a satellite or space-based system
• CPC Class Codes: H04W 24/00 Supervisory, monitoring or testing arrangements

Abstract

A voice quality performance evaluator includes an interface, memory, processing circuitry and may include a network interface. The interface is adapted to couple to a voice transmission path under test. The memory couples to the interface and is adapted to store an original voice sample and a modified voice sample. The processing circuitry couples to the interface and to the memory, retrieves the original voice sample from the memory and transmits the original voice sample to the voice transmission path under test via the interface. The processing circuitry also receives the modified voice sample via the interface from the voice transmission path under test. Finally, the processing circuitry compares the original voice sample to the modified voice sample in the frequency domain to evaluate the performance of the voice transmission path. In one construction, the voice quality performance evaluator includes a network interface that directs the communication network to complete the voice transmission path from and to the voice terminal. In one type of test that is performed by the voice quality performance evaluator, the voice transmission path includes only a vocoder of a voice terminal. However, in another type of test performed by the voice quality performance evaluator, the voice transmission path further includes a communication network. In one implementation, the original voice sample comprises a plurality of different voice samples. These plurality of different voice samples are separated by time markers. The time markers assist in allowing the voice quality performance evaluator to align the original voice sample and the modified voice sample prior to evaluating same. In performing its analysis, the voice quality performance evaluator divides the original voice sample and the modified voice sample into a plurality of time segments and performs an evaluation for each time segment. The voice quality performance evaluator then accumulates the evaluation for each time segment to produce an evaluation for the complete modified voice sample. Once the evaluation is complete, the voice quality performance evaluator may normalize the evaluation of the voice transmission path to other evaluations.

Claims

1. A voice quality performance evaluator employed to evaluate the performance of a voice transmission path, the voice quality performance evaluator comprising:

an interface adapted to couple to a voice transmission path under test;

memory coupled to the interface adapted to store an original voice sample, wherein the original voice sample includes a plurality of original voice sample segments;

processing circuitry coupled to the interface and to the memory that retrieves the original voice sample from the memory and that transmits the original voice sample on the voice transmission path under test via the interface;

the processing circuitry receiving a modified voice sample via the interface from the voice transmission path under test, wherein the modified voice sample includes a plurality of modified voice sample segments that correspond to the plurality of original voice sample segments;

the processing circuitry performing time alignment to align each of the plurality of original voice sample segments with its corresponding modified voice sample segments; and

the processing circuitry comparing the original voice sample to the modified voice sample in the frequency domain to evaluate the performance of the voice transmission path, wherein in comparing the original voice sample to the modified voice sample, the processing circuitry compares each original voice sample segment with its corresponding modified voice sample segment.

2. The voice quality performance evaluator of claim 1, wherein the voice transmission path comprises a vocoder of a voice terminal.

3. The voice quality performance evaluator of claim 2, wherein the voice transmission path further comprises a communication network.

4. The voice quality performance evaluator of claim 3, further comprising a network interface that directs the communication network to complete the voice transmission path from and to the voice terminal.

5. The voice quality performance evaluator of claim 1, wherein the plurality of original voice sample segments comprise actual voice samples taken from a plurality of speakers.

6. The voice quality performance evaluator of claim 5, wherein the plurality of original voice sample segments are separated by a plurality of time markers.

7. The voice quality performance evaluator of claim 1, wherein the processing circuitry performs Fourier transforms on the original voice sample and the modified voice sample.

8. The voice quality performance evaluator of claim 1, wherein:

each of the plurality of original voice sample segments is taken from a corresponding respective speaker; and

the plurality of voice sample segments have diverse speech characteristics.

9. The voice quality performance evaluator of claim 8, wherein the processing circuitry accumulates the evaluation for each of the plurality of original voice sample segments to produce an evaluation for the complete modified voice sample.

10. The voice quality performance evaluator of claim 1, wherein the processing circuitry normalizes the evaluation of the voice transmission path to other evaluations.

11. A method of evaluating the performance of a voice signal path, the method comprising:

transmitting an original voice sample onto the voice signal path, wherein the original voice sample includes a plurality of original voice sample segments;

receiving a modified voice sample from the voice signal path, the modified voice sample corresponding to the original voice signal, wherein the modified voice sample includes a plurality of modified voice sample segments that correspond to the plurality of original voice sample segments;

aligning in time each of the plurality of original voice sample segments with its corresponding modified voice sample segments;

transforming the plurality of modified voice sample segments and the plurality of original voice sample segments to the frequency domain; and

comparing the spectral components of the transformed plurality of modified voice sample segments to the spectral components of the transformed plurality of original voice sample segments to evaluate the performance of the voice transmission path.

12. The method of claim 11, wherein the voice transmission path passes through a vocoder of a voice terminal.

13. The method of claim 12, wherein the voice transmission path further passes through a communication network.

14. The method of claim 13, further comprising directing the communication network to complete the voice transmission path from and to the voice terminal.

15. The method of claim 11, wherein the plurality of original voice sample segments contain actual voice samples taken from a plurality of speakers.

16. The method of claim 15, wherein the plurality of original voice sample segments are separated by a plurality of time markers.

17. The method of claim 11, wherein transforming the modified voice sample and the original voice sample to the frequency domain includes performing Fourier transforms on the original voice sample and the modified voice sample.

18. The method of claim 11, wherein:

each of the plurality of original voice sample segments is taken from a corresponding respective speaker; and

the plurality of voice sample segments have diverse speech characteristics.

19. The method of claim 18, further comprising accumulating the evaluation for each of the plurality of original voice sample segments to produce an evaluation for the complete modified voice sample.

20. The method of claim 11, further comprising normalizing the evaluation of the voice transmission path to other evaluations.

21. The method of claim 11, further comprising windowing the modified voice sample and the original voice sample when transforming the modified voice sample and the original voice sample to the frequency domain.

22. The method of claim 11, wherein, in comparing the spectral components of the transformed modified voice sample to the spectral components of the transformed original voice sample to evaluate the performance of the voice transmission path, the difference in spectral properties is determined.