Data processing: speech signal processing – linguistics – language – Speech signal processing – Application
Reexamination Certificate
2000-11-30
2004-08-03
McFadden, Susan (Department: 2655)
Data processing: speech signal processing, linguistics, language
Speech signal processing
Application
C704S270000, C379S088010
Reexamination Certificate
active
06772123
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to the field of performing speech recognition. The present invention is directed to performing speech recognition for an internet appliance at a remotely located speech recognition application.
2. Related Art
In a technology that generates computers that are bigger, faster, more complicated and expensive, a growing market of low cost and specialized computer systems that are part of a larger networked system has emerged. One of the new categories of computer systems is the “internet appliance”.
An internet appliance is an embedded electronic system at the edge of the Internet. The internet appliance is inexpensive, simple to use, and usually designed to perform a dedicated or narrow range of functions. It maintains its low cost by minimizing hardware resources, including its memory capacity, processor power, and size. Some internet appliances are portable. Since these devices are connected to a network, such as the Internet, these internet appliances rely on a remote computer with its associated features to do any function that a computer may do, including storage of data and performance of an application.
The internet appliance focuses on one particular function to simplify operations, maximize convenience, and reduce cost. For example, an internet appliance may provide limited access to the Internet by featuring selected web sites, one of which could be a web site providing weather information. As selected by the user, the internet appliance might continually display updated weather information with data from the specific weather web site for any particular area without any further input from the user.
Internet appliances will eventually become more prevalent throughout the home or business in performing numerous tasks. Virtually any household appliance that can incorporate a digital interface can be an internet appliance. Complex functions such as heating and cooling of the home interior, answering and recording telephone messages, sprinkling the yard, securing the home with a security system, accessing the Internet, etc., or more simplistic functions such as making coffee or toast can be controlled using an internet appliance.
Previously, internet appliances had limited speech recognition capabilities.
FIG. 1
shows an electronic device
100
with speech recognition capabilities. Device
100
needs sophisticated hardware in order to run the algorithms needed for speech recognition. As such, device
100
is typically a stand-alone personal computer with a fast digital signal processor
115
to perform the speech recognition application
130
embedded within device
100
to maximize real time response times. Speech is inputted into voice recorder
110
in analog form. A converter
125
converts the analog signal to digital form whereupon the digital signal processor
115
operates the speech recognition application
130
to translate the input into text form that is stored in memory
120
and viewed at the display
105
.
Because of its simplistic construction and emphasis on reducing cost by limiting the hardware attached to it, an internet appliance would have limited capability to independently run a speech recognition application without excessively increasing the size, power, and cost of the internet appliance.
FIG. 2
shows a network architecture
200
for performing speech recognition that is explained by Z. Tu and P. Loizou in their paper, “Speech Recognition Over the Internet Using Java,” IEEE ICASSP 1999, Phoenix, Ariz. This architecture requires that the voice data is recorded at the appliance
205
and sent through a communication link
210
to the Internet Service Provider
220
, and then over the Internet
225
to the proxy server
230
. The proxy server
230
is pointed at a web site, or URL address, having a digital signal processor disposed therein that runs a speech recognition application. The speech recognition application translates the input voice data into text. Results from the speech recognition application are then sent back over the same communications network to the internet appliance
205
.
This network architecture
200
overcomes the problem of locating the speech recognition application along with the necessary hardware to run the application on the internet appliance
205
by utilizing a world wide web browser located on appliance
205
to do speech recognition via a remote proxy server
230
pointed at a web site with a speech recognition application disposed therein. However, the internet appliance requires a web browser which adds undesirable cost and complexity to the appliance. A further disadvantage to this architecture is the inconvenience of pointing the browser to the specific web site that contains the speech recognition application. Yet another disadvantage is the inherent losses and delays associated with the Internet architecture. Since the web site is located within the Internet, the end to end round trip response time will be affected by queuing delays within the network switches, available bandwidth, packet reassembly delays, and processing delays by the speech recognition application. These previous delay limitations may also result in losses of data.
Thus, a need exists to provide an efficient and cost-effective method and system for performing speech recognition for an Internet appliance.
SUMMARY OF THE INVENTION
The present invention provides a method and system for performing speech recognition for an internet appliance using a remotely located speech recognition application. Specifically, a method and system for performing speech recognition is provided such that the internet appliance can receive and execute sophisticated verbal commands. Additionally, the present invention allows for inexpensive and easy to operate internet appliances. Also, the present invention provides a method and system for performing speech recognition that is separate from an internet appliance. Moreover, the present invention provides a method and system for performing speech recognition for an internet appliance with real time response characteristics.
The present invention provides these advantages and others not specifically mentioned above but described in the sections to follow. Specifically, the present invention discloses a method and system for performing speech recognition for an internet appliance using a remotely located speech recognition application. The invention includes an internet appliance that is connected through a network with either a stand-alone computer or a server computer located at the Internet Service Provider. Verbal commands directed to an internet appliance are received as analog signals and converted to digital signals. The digital signals are remotely translated into a set of instructions by a dedicated hardware device or a software program that operates a speech recognition application at either a stand-alone computer or server computer located at an Internet Service Provider in a form recognizable by the internet appliance. The internet appliance receives and executes the translated instructions.
These and other objects and advantages of the present invention will no doubt become obvious to those of ordinary skill in the art after having read the following detailed description of the preferred embodiments which are illustrated in the various drawing figures.
REFERENCES:
patent: 6125284 (2000-09-01), Moore et al.
patent: 6157705 (2000-12-01), Perrone
patent: 6304573 (2001-10-01), Hicks, III
patent: 6327343 (2001-12-01), Epstein et al.
patent: 6334103 (2001-12-01), Surace et al.
patent: 6418199 (2002-07-01), Perrone
patent: 6477240 (2002-11-01), Lim et al.
patent: 6526335 (2003-02-01), Treyz et al.
patent: 6606611 (2003-08-01), Khan
Zhemin Tu, Philipos C. Loizou, “Speech Recognition Over the Internet Using Java”, 1999, IEEE 1CASSP.
Cooklev Todor
Gibbs Darrin
Gray Mark
Morley Ken
3Com Corporation
McFadden Susan
LandOfFree
Method and system for performing speech recognition for an... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and system for performing speech recognition for an..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and system for performing speech recognition for an... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3305109