Data processing: speech signal processing – linguistics – language – Speech signal processing – Application
Reexamination Certificate
1999-03-15
2003-11-04
Knepper, David D. (Department: 2654)
Data processing: speech signal processing, linguistics, language
Speech signal processing
Application
C704S257000
Reexamination Certificate
active
06643620
ABSTRACT:
BACKGROUND AND SUMMARY OF THE INVENTION
The present invention is directed to a voice controlled system for recording and retrieving audio/video programs. More particularly, the present invention is directed to a voice controlled multimedia system for receiving and processing spoken requests against a multimedia database comprising electronic programming guide information for recording and retrieving the audio/video programs.
The next generation televisions and related accessories (set-top box, VCR, audio/video processor, satellite or cable receiver, etc.) will have significant processing power made available by a CPU or DSP. This processing power can be used to support tasks which are very different from what the device was originally intended for (mainly decoding and processing the video and audio signals), so that the unit can be enhanced with various functions at little or no cost for the manufacturer.
However, systems which utilize a voice activated controller for programming a multimedia database are conspicuously absent from the prior art. For example, in U.S. Pat. No. 5,293,357, a method is described for programming an event timer and recording television broadcasts by using an on-line TV schedule listing. The user manually selects the desired program from the on-line listings, and the selection is translated into an event for the timer.
In the present invention, information collected from an electronic programming guide (EPG) or entered by the user, is stored in a program database. The user can then retrieve programs by providing a natural language description of what he or she desires to play back. The recording request programming step can also be accomplished by giving a description in natural language of the desired program to be recorded. Furthermore, the user can program an event even if it is not listed in the EPG available at that time, because the present invention will keep updating the EPG (for example, on weekly or monthly basis) and try to resolve recording requests that are still pending. Another advantage of the present invention is that it can monitor the EPG for a particular set of programs indefinitely. For example, a sports fan can give a complex command like “record all the basketball games featuring the L.A. Lakers,” and he or she will be able to record all Lakers game.
In U.S. Pat. No. 4,873,584, a system is described in which a computer controls a VCR and provides means for storing a list of the television programs recorded by the VCR. The system also provides means for playing back the programs on the VCR in any preferred order. However, this system also requires the user to manually enter the recording and play back requests.
In the present invention a computer is not needed, and the microprocessor present in a set-top box or an A/V decoder can be used to perform all the functions. In addition, the program schedule listings do not need to be recorded on a floppy disk but can be obtained from a TV channel or from an internet or telephone connection. The device of the present invention can thus be programmed for a potentially unlimited period of time, instead of a week at a time. The present invention also provides means for automatically maintaining a database of the available programs and for retrieving titles using natural language spoken requests and commands.
In U.S. Pat. No. 5,475,835, a computer controls an A/V player/recorder and provides functions for maintaining a home entertainment media inventory. This device uses infrared communication between the computer and the player/recorder. The computer interface is provided by a series of touch screen menus which can be used for controlling and programming the A/V devices. However, the computer does not provide an interface which can accept programming commands in a natural language format.
In the present invention a dedicated computer is not needed, nor is the user required to operate the computer to retrieve programs. Commands presented to the device of the present invention can be given using naturally spoken language and can perform complex operations. A dialogue system can intervene to resolve ambiguities or to prompt the user for additional information.
In view of the foregoing, it is desirable to provide a system which can understand spoken requests and process the user's request against a multimedia database of records. It is further desirable to receive a spoken request to record a desired program and provide a system for searching for the airing time of the requested program in a database of electronic programming guide records. It is also desirable to provide a system which allows a library of multimedia programs to be maintained in the multimedia database and present the system with a spoken request to retrieve a title from the multimedia database. Finally it is desirable to allow the user to update the library of multimedia programs using spoken natural language requests and commands.
The present invention provides a voice controlled system for recording audio/video (A/V) programs using a VCR, DVD or video disc recording device, or any other device capable of storing A/V streams. The present invention also provides a system for retrieving programs from tape, DVD, CD, or any other device capable of playing back A/V media using spoken natural language requests and commands. The invention can also maintain a database of the programs available in a personal multimedia library and provide typical database associated functions such as information retrieval, statistics, and cataloging.
The invention also provides a technique for generating recording requests and building the information and program records in the multimedia database either manually or automatically. Information can be entered manually using an input device (optical reader, by selecting text with a remote control, etc.) or by voice, and then converted into text by a speech recognition system. Information and program records can also be extracted automatically from an electronic program guide (EPG) and can consist in the title, author(s), player(s), summary, description of the program, or any combination thereof. Text information can be classified into two categories: the stored program records which are searched by the system for allowing the user to record A/V programs, and the information used to retrieve A/V programs.
The records forming the multimedia database are stored in a memory device including but not limited to static RAM or a magnetic storage device, and contain a code that uniquely identifies the media (video tape, CD, DVD disk, etc.) and the location of the program within the media (tape position, CD track, etc.). The text within the records can be used to dynamically generate a vocabulary (eventually completed by additional words) utilized by a natural language processor, so that a user can give a spoken, natural language description of the desired program to record retrieve. After processing and understanding the spoken request, the system will record or play back the program that most closely matches the description. If the media is not currently loaded in the playback device (VCR, multi-disc DVD player, etc.), the system of the present invention will provide the user with a way to identify the appropriate media (tape catalog number, title, etc.) and ask the user to load the requested media. The system will then position the media to the desired program and commence playback.
Similarly, the information associated with programs in an EPG can be used for the purpose of selecting a program for unattended recording by an appropriate video recording device. For example, the user gives a spoken description of the desired program which is then converted into text by a speech recognizer. When a program is found in the EPG that matches the description, it is scheduled for recording. In this way, an indefinite period of time can be monitored for recording by the system of the present invention, even if the EPG has a limit time coverage. For example, the user may request something like: “Please record the mov
Contolini Matteo
Junqua Jean-Claude
Kuhn Roland
Harness Dickey & Pierce PLC
Knepper David D.
Matsushita Electric - Industrial Co., Ltd.
LandOfFree
Voice activated controller for recording and retrieving... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Voice activated controller for recording and retrieving..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Voice activated controller for recording and retrieving... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3165634