Telecommunication system tester with voice recognition capability

• US 5,572,570 A
• Filed: 10/11/1994
• Issued: 11/05/1996
• Est. Priority Date: 10/11/1994
• Status: Expired due to Term

• Current Assignee: Empirix Incorporated (Empirix Holdings I, Inc.), Transaction Environments Incorporated
• Original Assignee: Teradyne Incorporated
• Inventors: Kuenzig, John F.
• Primary Examiner(s): Chin, Wellington
• Assistant Examiner(s): Shankar, Vijay
• Application Number:
  US08/321,357
• Time in Patent Office: 756 Days
• Field of Search: 379/1, 379/9, 379/10, 379/12, 379/18, 379/14-15, 379/69, 379/88, 379/89, 379/27, 379/29, 379/32
• US Class Current: 379/1.02
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: H04M 3/24 with provision for checking the normal operation
• CPC Class Codes: H04M 3/323 for the arrangements providing the connection (test connection, test call, call simulation)
• CPC Class Codes: H04M 3/487 Arrangements for providing information services, e.g. recorded voice services or time announcements

Abstract

Connection is made directly or via a switched telephone network between a tester and a telecommunication system to be tested. User data, such as voice signals, and signalling data, such as DTMF signals are generated by the tester to cause the telecommunication system to respond. The user and signalling data may be generated interactively to test the telecommunication system in real time, or by executing prerecorded scripts. When the tester is used interactively, the representations of the signals transmitted to and from the tester are recorded in at least one script file. Voice recognition is preferably used to determine whether the response received from the telecommunication system is the expected response, particularly when a script is executed. The vocabulary used for voice recognition may be created by repeatedly executing a script file (recorded during interactive operation of the tester or created manually) which causes the telecommunication system to generate the desired prompt(s) and message(s). Alternatively, the vocabulary may be obtained in other ways, such as by reproducing voice signals at the telecommunication system.

Claims

1. A method of testing a telecommunication system, connectable to a telecommunication network, comprising the steps of:

(a) generating, in a data processing device, test signals simulating at least one of voice and non-voice signals transmissible by a telecommunication network;

(b) transmitting the test signals to the telecommunication system;

(c) detecting response signals produced by the telecommunication system in response to the test signals, said response signals including initial voice response signals generated automatically by the telecommunication system;

(d) recording at least one sampler script including representations of only the test signals generated in step (a);

(e) recording at least one recognizer script including representations of booth the test signals generated in step (a) and the response signals detected in step (c);

(f) repeatedly reproducing the test signals using the at least one sampler script recorded in step (d);

(g) reproducing the test signals using the at least one recognizer script recorded in step (e);

(h) detecting subsequent voice response signals produced by the telecommunication system in response to the test signals reproduced in step (f);

(i) recording the subsequent voice response signals, wherein said recording in step (i) is repeated to obtain a plurality of voice response files, at least one of the voice response files obtained for each repetition of the test signals reproduced in step (f);

(j) building a vocabulary file, wherein said building of the vocabulary file in step (j) uses the voice responses files recorded in step (i);

(k) detecting subsequent voice response signals produced by the telecommunication system in response to the test signals reproduced in step (g); and

(l) performing automatic voice recognition on the subsequent voice response signals detected in step (k) using the vocabulary file built in step (j), wherein said detecting in step (k) and the automatic voice recognition in step (l) are performed.

2. A method as recited in claim 1,

wherein said reproducing in step (g), said detecting in step (k), and the automatic voice recognition in step (l) are performed repeatedly while the telecommunication system is available to serve calls from users.

3. A method of testing a telecommunication system, connectable to a telecommunication network, comprising the steps of:

(a) generating, in a data processing device, test signals simulating at least one of voice and non-voice signals transmissible by a telecommunication network;

(b) transmitting the test signals to the telecommunication system;

(c) detecting response signals produced by the telecommunication system in response to the test signals;

(d) recording at least one sampler script including representations of only the test signals generated in step (a);

(e) recording at least one recognizer script including representations of both the test signals generated in step (a) and the response signals detected in step (c);

(f) executing the at least one sampler script to control the substeps of

(f1) reproducing the test signals recorded in step (d), and

(f2) recording initial voice response signals generated in response to the test signals reproduced in step (f1); and

(g) executing the at least one recognizer script to control the substeps of

(g1) reproducing the test signals recorded in step (e),

(g2) detecting subsequent voice response signals generated in response to the test signals reproduced in step (g1), and

(g3) automatically performing voice recognition on the subsequent voice response signals detected in step (g2) based on the initial voice response signals recorded in step (f2).

4. A tester for a telecommunication system connectable to a telecommunication network, comprising:

at least one processor controlling generation of test signals simulating at least one of voice and non-voice signals transmissible by the telecommunication network;

at least one interface, coupling said processor to the telecommunication system, to transmit the test signals and to receive response signals generated by the telecommunication system in response to the test signals, said response signals including voice response signals;

at least one storage device, coupled to at least said processor, storing the test and response signals; and

at least one speech recognition device, coupled to at least said processor, for analyzing the voice response signals.

7. A tester as recited in claim 4,

wherein said interface includes an analog connection, a call driver connection, a ringdown connection, and a digital connection, thereby allowing the tester to operate with or without connection to a telecommunication network.

5. A method of testing a telecommunication system including a voice response unit using a data processing device, comprising the steps of:

(a) receiving commands from a human operator of the data processing device to test the telecommunication system in real-time;

(b) generating, in the data processing device, test signals simulating user data and signaling data;

(c) transmitting the test signals to the telecommunication system;

(d) recording representations of the test signals in a sampler and recognizer scripts in temporary storage on the data processing device;

(e) storing representations of response signals produced by the telecommunication system in the recognizer script, including machine produced voice signals from the voice response unit and instructions to perform speech recognition on the machine produced voice signals;

(f) repeating steps (a)-(e) until a command is received from the human operator to step;

(g) prompting the human operator to save the sampler and recognizer scripts;

(h) storing each of the sampler and recognizer scripts in separate permanent files when instructed by the human operator in response to said prompting in step (g);

(i) repeatedly reproducing the sampler script to obtain a plurality of vocabulary samples of the machine produce voice signals from the voice response unit;

(j) automatically building a vocabulary of words from the samples obtained during said reproducing in step (i); and

(k) reproducing the recognizer script to obtain at least one test sample of the machine produced voice signals from the voice response unit and to automatically perform speech recognition on the test sample using the vocabulary built in step (j).

6. A method as recited in claim 5, further comprising the steps of:

(l) connecting the data processing device to the telecommunication system via a switched telephone network in response to the commands received in step (a); and

(m) repeatedly performing said reproducing in step (k) during operation of the telecommunication system to serve users.

8. A method of testing a telecommunication system, the system having a speech interface and being connectable to a telecommunication network, comprising the steps of:

(a) receiving voice signals produced by the telecommunication system;

(b) transmitting test signals to the telecommunication system; and

(c) evaluating voice response signals using the voice signals received in step (a), said voice response signals produced by the telecommunication system in response to the test signals transmitted in step (b), thereby determining whether the telecommunication system is operating properly.

9. A method as recited in claim 8,

wherein said telecommunication network is a public switched telephone network.

10. A method as recited in claim 8,

wherein said receiving in step (a) includes the substeps of:

(a1) adding representations of said voice signals to a vocabulary file, and

(a2) optionally editing the vocabulary file.

11. A method as recited in claim 10,

wherein said editing in step (a2) includes the substeps of:

(i) listening to at least one of said representations of voice signals in the vocabulary file, and

(ii) optionally deleting the representations of voice signals listened to in step (i).

12. A method as recited in claim 10,

wherein said evaluating in step (c) includes the substeps of:

(c1) comparing said voice response signals with said representations of voice signals in the vocabulary file to determine whether the voice response signals match any of the representations in the vocabulary file,

(c2) comparing the representations that match the voice response signals with expected response data to determine whether the telecommunication system produced the proper voice response signals, and

(c3) indicating whether the telecommunication system produced the proper voice response signals in response to the test signals transmitted in step (b).

13. A method of testing a telecommunication system, the system being connectable to a telecommunication network, comprising the steps of:

(a) interactively transmitting test signals to the telecommunication system and detecting initial response signals produced by the telecommunication system;

(b) automatically recording representations of both the test signals transmitted and the initial response signals detected in step (a);

(c) reproducing the test signals recorded in step (b);

(d) evaluating subsequent response signals produced by the telecommunication system in response to the test signals reproduced in step (c) to determine whether the telecommunication system is operating properly.

14. A method as recited in claim 13,

wherein said initial response signals include voice response signals, and

wherein said interactively transmitting and detecting in step (a) is performed by a human operator who can hear the voice response signals.

15. A method as recited in claim 13,

wherein said interactively transmitting and detecting in step (a) comprises the substeps of:

(a1) transmitting test signals to the telecommunication system,

(a2) detecting initial response signals produced by the telecommunication system,

(a3) transmitting test signals to the telecommunication system based on the initial response signals detected in step (a2), and

(a4) repeating steps (a2) and (a3) until the telecommunication system is fully tested.

16. A method as recited in claim 13,

wherein said recording in step (b) comprises the substeps of:

(b1) automatically recording at least one file including representations of only the test signals transmitted in step (a), and

(b2) automatically recording at least one file including representations of both the test signals transmitted and the initial response signals detected in step (a).

17. A method as recited in claim 16,

wherein said evaluating in step (d) comprises the substeps of:

(d1) comparing said subsequent response signals with said representations of initial response signals recorded in step (b2), to determine whether the telecommunication system produced the proper response signals, and

(d2) indicating whether the telecommunication system produced the proper response signals in response to the test signals reproduced in step (c).

18. A method as recited in claim 16,

wherein said reproducing in step (c) comprises the substep of:

(c1) reproducing the test signals recorded in step (b1).

19. A method as recited in claim 18,

wherein said recording in step (b) further comprises the substep of:

(b3) automatically adding representations of subsequent response signals to the file recorded in step (b2), said subsequent response signals produced in response to the test signals reproduced in step (c1).

20. A method as recited in claim 16,

wherein said reproducing in step (c) comprises the substep of:

(c1) reproducing the test signals recorded in step (b2).

21. A method as recited in claim 16,

wherein said recording in steps (b1) and (b2) further includes automatically recording representations of signaling data, timing information, and test status information.

22. A method as recited in claim 21,

wherein said representations of signaling data include DTMF signals, MF signals, and indications of changes in polarity.

23. A method as recited in claim 21,

wherein said representations of timing information include expected response times, said expected response times including the time when a response signal is expected relative to the time when a test signal is transmitted.

24. A method as recited in claim 21,

wherein said representations of test status information include collection type, said collection type selected from the group consisting of DTMF signals, MF signals, and voice signals.

25. A method as recited in claim 24,

wherein said collection type is DTMF signals,

wherein said recording in step (b1) further includes recording a first instruction to collect DTMF signals from the telecommunication system, and

wherein said recording in step (b2) further includes recording said first instruction to collect DTMF signals from the telecommunication system, recording a second instruction to save representations of the DTMF signals in said file, and recording a third instruction to compare said representations of the DTMF signals with the representations of the initial response signals.

26. A method as recited in claim 24,

wherein said collection type is MF signals,

wherein said recording in step (b1) further includes recording a first instruction to collect MF signals from the telecommunication system, and

wherein said recording in step (b2) further includes recording said first instruction to collect MF signals from the telecommunication system, recording a second instruction to save representations of the MF signals in said file, and recording a third instruction to compare said representations of the MF signals with the representations of the initial response signals.

27. A method as recited in claim 24,

wherein said collection type is voice signals,

wherein said recording in step (b1) further includes recording a first instruction to collect voice signals from the telecommunication system, and

wherein said recording in step (b2) further includes recording said first instruction to collect voice signals from the telecommunication system, recording a second instruction to save representations of the voice signals in said file, recording a third instruction to reserve a speech recognition device, recording a fourth instruction to direct said speech recognition device to analyze the representations of the voice signals, and recording a fifth instruction to compare said representations of the voice signals with the representations of the initial response signals.

28. A method of testing a telecommunication system using a data processing device, the system being connectable to a telecommunication network, comprising the steps of:

(a) connecting the data processing device to the telecommunication system, without connection through a telecommunication network;

(b) interactively transmitting test signals to the telecommunication system and detecting response signals produced by the telecommunication system, said test signals simulating user data and signaling data; and

(c) evaluating the response signals to determine whether the telecommunication system is operating properly.

29. A method as recited in claim 28,

wherein said interactively transmitting and detecting in step (b) comprises the substeps of:

(b1) transmitting test signals to the telecommunication system,

(b2) detecting response signals produced by the telecommunication system,

(b3) transmitting test signals to the telecommunication system based on the response signals detected in step (b2), and

(b4) repeating steps (b2) and (b3) until the telecommunication system is fully tested.

30. A method as recited in claim 28,

wherein said evaluating in step (c) comprises the substeps of:

(c1) comparing the response signals detected in step (b) with expected response signals, to determine whether the telecommunication system produced the proper response signals, and

(c2) indicating whether the telecommunication system produced the proper response signals in response to the test signals transmitted in step (b).