Telephonic communications – Diagnostic testing – malfunction indication – or electrical... – Testing of network terminating interface – subscriber trunk...
Reexamination Certificate
1999-04-09
2003-01-07
Tsang, Fan (Department: 2645)
Telephonic communications
Diagnostic testing, malfunction indication, or electrical...
Testing of network terminating interface, subscriber trunk...
C379S027040, C379S088080, C379S001020, C379S010010, C375S224000
Reexamination Certificate
active
06504905
ABSTRACT:
TECHNICAL FIELD
This invention relates to a system and method for testing a voice interface of a telecommunication system, particularly a voice messaging system (VMS).
BACKGROUND ART
Testing of telecommunication systems involves evaluating the connections and signal quality of the system as well as the performance of services provided by the system. In particular, many telecommunication systems provide a voice messaging system (VMS) which must be tested to ensure that it is functioning properly, especially at the user interface.
In order to test a VMS, a testing system must provide the normal inputs expected by the VMS, as well as monitor and recognize all the possible responses of the VMS. The inputs to a VMS can include both voice signals and non-voice signals, such as dual tone multi-frequency (DTMF) signals, used by a caller to interact with and navigate through the VMS. The responses of the VMS typically include voice prompts to the caller for the entry of voice or non-voice data. It is these voice prompts which define the user interface of the VMS, and which must be tested for quality and accuracy.
Typically, the multitude of voice prompts from the VMS are tested by comparing them with an equivalent number of reference prompts stored by the testing system. In the past, voice recognition has been executed by recognizing and comparing voice prompts word by word, by selected words, or by segmenting words into syllables or phonemes. However, these previous methods require substantial memory in order to store the myriad of different inputs and responses for the VMS, as well as considerable processor capabilities for making the prompt comparisons. Furthermore, such previous methods are time-consuming and inflexible in their manner of prompt recognition.
DISCLOSURE OF INVENTION
Therefore, it is an object of the present invention to provide a system and method of testing voice signals in a telecommunication system that uses phrases of voice prompts to execute voice recognition.
It is a further object of the present invention to provide a system and method of testing voice signals in a telecommunication system that requires less information to be stored to accomplish voice recognition.
It is a still further object of the present invention to provide a system and method of testing voice signals in a telecommunication system that decreases the time required for voice recognition while maintaining accuracy.
Accordingly, a method of testing voice signals in a telecommunication system is provided. The method includes generating a test signal to elicit a response from the telecommunication system and detecting a voice signal produced by the telecommunication system in response to the test signal. The method further includes accessing stored reference phrases, wherein one or more designated reference phrases can be combined to form an expected voice signal. Still further, the method includes comparing the detected voice signal with the designated reference phrases to determine if the detected voice signal substantially corresponds to the expected voice signal.
In a preferred embodiment, the step of comparing the detected voice signals with the designated reference phrases first includes trimming leading silence from the voice signal and determining the average acoustic intensity of the voice signal. Next, the voice signal is divided into component frames of substantially equal time durations. Then, the average acoustic intensity and the number of zero crossings for each voice signal frame is determined, after which the average acoustic intensity of each voice signal frame is normalized by the average acoustic intensity of the entire voice signal. The preferred comparison method further includes comparing each voice signal frame with a corresponding reference phrase frame of the same time interval. Once all of the corresponding frames have been compared, the matching percentage between the voice signal frames and the reference phrase frames is determined. Still further, the preferred comparison method includes comparing the matching percentage to a predetermined threshold to determine if the reference phrase is actually present in the voice signal.
Correspondingly, a system of testing voice signals in a telecommunication system is provided. The system includes a processor for controlling the generation of a test signal to elicit a response from the telecommunication network. The system further includes a telecommunication interface in communication with the processor and the telecommunication system to generate the test signal and to detect a voice signal produced by the telecommunication system in response to the test signal. A database in communication with the processor stores the detected voice signal and a plurality of reference phrases, wherein one or more designated reference phrases can be combined to form an expected voice signal. Voice recognition software operable with the processor then compares the detected voice signal with the designated reference phrases to determine if the detected voice signal substantially corresponds to the expected voice signal.
The above objects and other objects, features, and advantages of the present invention are more readily understood from a review of the attached drawings and the accompanying specification and claims.
REFERENCES:
patent: 5091948 (1992-02-01), Kametani
patent: 5536902 (1996-07-01), Serra et al.
patent: 5572570 (1996-11-01), Kuenzig
patent: 5633909 (1997-05-01), Fitch
patent: 5710803 (1998-01-01), Kowal et al.
patent: 5797123 (1998-08-01), Chou et al.
patent: 5822397 (1998-10-01), Newman
patent: 5825855 (1998-10-01), Astarabadi
patent: 5835565 (1998-11-01), Smith et al.
patent: 5850627 (1998-12-01), Gould et al.
patent: 5933475 (1999-08-01), Coleman
Tsai Louis K.
Wu Ying-li
Brooks & Kushman P.C.
Qwest Communications International Inc.
Sing Simon P.
Tsang Fan
LandOfFree
System and method of testing voice signals in a... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method of testing voice signals in a..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method of testing voice signals in a... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3004842