Telephonic communications – Audio message storage – retrieval – or synthesis – Voice activation or recognition
Reexamination Certificate
2000-09-02
2004-03-02
Tsang, Fan (Department: 2645)
Telephonic communications
Audio message storage, retrieval, or synthesis
Voice activation or recognition
C379S088020, C370S249000, C455S423000, C704S233000
Reexamination Certificate
active
06700953
ABSTRACT:
CROSS-REFERENCE TO RELATED APPLICATIONS
“Not Applicable”
STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT
“Not Applicable”
REFERENCE OF A “MICROFICHE APPENDIX”
“Not Applicable”
BACKGROUND
1. Field of Invention
The present invention relates to systems analysis and more particularly, to a system, apparatus, method and article of manufacture for evaluating the quality of a transmission channel using voice recognition technology. Among other features to be described in more detail later, the system, apparatus, method and article of manufacture is particularly suited for providing an objective measurement, in real-time, of the quality of audio signals transmitted along the transmission channel.
2. Description of the Related Art
Many technologies are used in today's telecommunication systems, so powerful test tools are essential for designing, building and optimizing such systems. Engineers require tools that measure technology-specific information as well as overall quality of service. In industries such as wireless telecommunication and Voice Over IP (VoIP), intense competition has driven companies to provide better and better quality at lower end-user cost. Test tools are needed that can objectively measure the quality of service (QoS) for differing telecommunication technologies while providing a common benchmark criteria for all.
The best way of measuring telecommunication QoS is to measure the actual quality of transmitted voice. Voice quality measurement can be subjectively measured using a panel of people but this can result in inconsistent results that are not repeatable. This method can also be expensive when used for subsequent trials, such as when engineers may take the results of the panel test, modify the network for better performance, and need to repeat the testing cycle. Engineers need a way of objectively measuring voice quality that is accurate, repeatable, economical, and consistent across different telecommunication technologies.
Telecommunication is international. Since many telecommunication vocoders are often designed to the operate for a particular language or voice type, QoS might digress if used by an international user base. Voice quality measurement tools must be the operational for different languages and voice types.
The first solutions for measuring voice quality were techniques of mean opinion score (MOS). Panels of people were used in a controlled environment to give an opinion of samples of voice quality. The scores from each person were then averaged to give a resulting MOS score. This solution was not repeatable, costly, and not practical for engineers designing telecommunication systems. It also required time to gather the panel of people, administer the tests, and formulate the resultant scores.
Early analog telecommunication systems would send audio tones across an audio channel to measure the received signal, noise and distortion (SINAD). Although this method was objective and repeatable, it became obsolete with the advent of digital communications technologies. Digital communications implement vocoders that digitally represent voice according to a set of voice characteristics. Since SINAD uses audio tones as the source, vocoders distort the signal and produce erroneous results.
Modern voice quality measurement solutions use the Perceptual Voice Quality Measure (PSQM) algorithm defined in ITU Recommendation P.861. Although the PSQM algorithm is accurate for many telecommunication technologies, it has some shortcomings that make it difficult to implement. Tools that use the PSQM algorithm must be calibrated using MOS testing. Since MOS testing is timely and expensive, it is difficult for test tool manufacturers using PSQM to quickly react to market needs. Furthermore, PSQM is difficult to implement in VoIP systems because large amounts of latency disrupt the synchronization of the received signal to the undistorted reference signal.
SUMMARY OF THE INVENTION
The present invention, sometimes referred to as Opera™, uniquely adapts voice recognition technology for the purpose of measuring the voice quality of telecommunication systems. With the rapid development of telecommunication technology, engineers require powerful test tools that can measure the performance of their system and compare that performance to other competing systems. The present invention provides an accurate practical tool that measures voice quality in real-time and records audio for later examination. Because objective voice quality scores are universal measurements that apply to all telecommunication systems, The present invention can be used for a wide variety of testing applications.
The present invention measures the voice quality of telecommunication systems in a way that is accurate, repeatable, and practical for real-time implementation. The present invention resolves the shortcomings of previous methods while providing the flexibility for different languages and different technologies. The present invention can also be used on high latency technologies such as those implemented with VoIP.
The key to the present invention's flexibility involves a unique application of voice recognition technology for the purpose of measuring voice quality. Voice recognition combined with high quality audio equipment and some custom software development creates a powerful and effective solution.
Therefore, in accordance with one aspect of the present invention there is provided a method for evaluating the quality of a transmission channel having a first port and a second port, comprising the steps of: transmitting a first signal from said first port to said second port via said transmission channel; receiving said first signal at said second port; analyzing said first signal using voice recognition technology to generate a quality score within a predetermined time frame; monitoring step (c) to ensure that said quality score is generated during said predetermined time frame; assigning a default quality score if a quality score is not generated during said predetermined time frame; transmitting a second signal from said second port to said first port, via said transmission channel, after said quality score is either generated or assigned said default quality score; and monitoring step (f) to ensure that said second signal is not being transmitted to said first port while said first signal is being processed.
In accordance with a second aspect of the present invention, there is provided a method further comprising the step of storing, in a storage medium, the first signal and the quality score.
In accordance with a third aspect of the present invention, there is provided a method further comprising the step of displaying said quality score.
In accordance with a third aspect of the present invention, there is provided a method wherein said quality score is displayed in real-time.
REFERENCES:
patent: 5835565 (1998-11-01), Smith et al.
patent: 5987320 (1999-11-01), Bobick
patent: 6275797 (2001-08-01), Randic
patent: 6304634 (2001-10-01), Hollier et al.
Elsasser Curtis Fred
Maurer Brett Pierre
Schechter Michael Jacob
Foley & Lardner
Gauthier Gerald
Metatron Technologies, Inc.
LandOfFree
System, apparatus, method and article of manufacture for... 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, apparatus, method and article of manufacture for..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System, apparatus, method and article of manufacture for... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3218040