SE502013C2 - Speech-controlled input device with standard logical interface - Google Patents

Abstract

The invention concerns a device with logic standard interface (11), for speech-controlled input of data into a computer, etc. (9). The device completely replaces a conventional input device, e.g., a keyboard and/or a control unit. The invention has been designed along the lines of speech-recognition technology and can either be a physically independent unit (5) or a logic unit (8), that is, a part of another device. A physically independent unit can consist of a physical interface for speech input (12), an audio card (12), a system program (3), a microcomputer (4), a physical interface for output (13), also a communications medium (6, 7) between the device and equipment for speech input (1) as well as the device and equipment that is to be operated (9). The invention is chiefly used for speech-control and hands-free use of equipment that has any kind of standard input device and/or control unit, e.g., a computer or a computer system.

Description

0 2 Û 1 3 2 Where TIA uses, for example, the code system according to the ANSI standard, then TIA is functionally exactly a keyboard and its drivers, but instead of pressing a physical keyboard, you speak into a microphone.

Such a speech-controlled device does not exist. (refer to news review report Dnr P 914/93 from the Patent Office).

All existing known speech recognition products that can generate text have the disadvantage that they do not have a standard interface, so they work with only one or a few applications. They are based on word recognition, which means that the product becomes very demanding, has limited vocabulary, has mare registers, is relatively slow, has little reliability, is speaker dependent, is a one-user system, has little tolerance for pronunciation changes, is sensitive to background noise, requires a lot of time for training the computer in order for it to "understand" the user's words and command, is difficult to learn and use and that it is relatively expensive.

Examples of such products are Voice type and Dictation systems from IBM and Dragon Dictate from Dragon Systems.

Another application that is based on the same principle and uses speech recognition technology is "Text editor for speech input", with patent number US, 4,914,704, hereinafter referred to as USP.

TIA differs totally in the basis, purpose and result of the device, from the referenced patent, USP.

So you can only compare TIA with a keyboard or mouse and their drivers but not with an application program, for example the word processor "Microsoft WORD for Windows".

In the same way, you can compare the word processor "Text editor for speech input" according to USP, with other word processors such as Word Perfect, Microsoft WORD for Windows, Dragon Dictate etc. but not with a keyboard driver.

In other words, just as it is not possible to compare the word processor Microsoft WORD for Windows with the spreadsheet program EXCEL or the fax program WINFAX, etc., it is also not possible to compare TIA with a word processor, for example "Text editor for speech input". 502 013 3 USP is a "text editor for speech input", ie. an independent application program that uses speech recognition technology to enter text into its own application, the text editor.

The text that is entered through the voice input into the "text editor" is not a standard code, but they use a so-called token and string format (USP: fi g. 3, ñg. 4), which are defined by the designer of the "text editor" for this particular application.

Therefore, this word processor has no interface to applications other than its text editor, so it cannot enter text into or control any application other than itself.

TIA, on the other hand, is a driver and uses only standard code according to a standard code system, to be able to completely replace a keyboard.

M.a.o. For example, according to the ANSI standard, the code “97” is sent instead of “a” and the code “00; 59” instead of “F1” as output, so that TIA will be a standard interface to all standard applications in that computing environment.

The USP only writes the writable characters to the text editor, which are included as part of the codes according to the ASCII table.

However, in addition to the default code for typing characters, TIA sends the default code for someone or a combination of some of all the keyboard's keys, according to those in a standard code system. These can be, for example: - Control and special keys eg: ESC, TAB, HOME, INSERT; - CTRL + character key ex: CTRL + A, CT RL + B; - ALT + character key ex: ALT + A, ALT + B; - SHIFT + tangent ex: SHIFT + TAB; - Function keys: F1, F2; - and any or a combination of any of these keys ex: CTRL + F1, ALT + F2 O.S.V.

Exactly the property that TIA has, ie. according to the characterizing parts of the claims, makes the TIA completely unique and has many unique features: 502 015 4 A: When the TIA uses the code system according to the ANSI standard, it becomes completely equivalent and behaves like a keyboard and its drivers. By this is meant that: A. 1: Unlike USP which enters text to its word processor, TIA is a general input device for controlling and inputting text to other applications.

A.2: TIA activates menus in menu-driven programs, to open the "blinds" and execute various commands found in submenus and dialog boxes, by sending out default codes for ALT + character key combinations. For example. The ALT + A combination has the standard code "O0; 30", which activates, for example, the "File" menu on the menu bar of any standard application that has an underlined A, "AÉ in its menu bar. This is just like pressing the ALT key and the A key at the same time on a regular computer keyboard.

A3: Using CT RL + character or other key, TIA activates short commands, ie. activate commands without having to open the blinds, and perform the defined functions of these codes in various programs.

A.4: TIA also sends the code for control and special keys such as ESC, HOME, Function keys and the code for the other key combinations according to the standardized code system.

A.5: TIA controls the cursor on the screen by using Escape sequences according to the ANSI standard.

USP does not perform any of these at all, except that it can write only writable characters, ie. some of the ASCII characters, by their own code format.

B: To be able to apply TIA to other devices that have a keyboard and use a standard code or signaling system, TIA uses to combine standardized code or signaling systems. In this way, new products are created that are controlled by voice instead of with a keyboard. 502 013 5 C: In addition, TIA, p.g.a. its construction, a device which, among other things: - easy and quick to install and get started with, - easy to learn and easy to use, - can be very easily translated into other languages, - has "built-in" unlimited vocabulary.

- HLIII.

Where e.g. ASCII code is used as the default code, the device can, in addition to generating the ASCII code for numbers and letters, also generate the code for other characters, e.g. control characters used for control and communication purposes, graphic, foreign and mathematical symbols. This feature allows the system to automatically have a built-in "unlimited" vocabulary. In addition, p.g.a. that a set of the Extended ASCII Code, Extended ASCII Code, also contains character codes for foreign characters, the device has a built-in unlimited "multilingual" vocabulary.

The invention is a simple, flexible and flexible product from the user's point of view, as it eliminates or reduces the above-mentioned disadvantages and problems of word-based speech recognition products. In addition, this invention has some other features that do not exist in other systems, which open up new possibilities, areas of use and product ideas.

The means necessary to achieve such a device are a device for introducing voice signals Lex. microphone, sound card, system driver, ie. TIA's program, microcomputer for the physically independent input device and a communication medium.

An embodiment of the invention is shown schematically in the accompanying drawings.

Figure 1 shows an illustration of principles.

Figure 2 shows a physically independent speech-controlled input device and 3 shows a logical one.

Figure 4 shows where application programs and TIA's programs are located in a computer system.

TIA's program, which a system operates, is a standard interface between an application program and an external device, such as a microphone. 502 015 6 TIA can be designed with either speaker-dependent or speaker-independent recognition, depending on the application in which the TIA is to be used. If user identification and restriction of use of the device is desired, a speaker-dependent recognition must be used and if e.g. a Her-user system is desired, then a speaker-independent recognition is preferred.

According to Figs. 2 and 3, the TIA consists of hardware and software, which are mainly: a device for introducing voice signals into the system, e.g. microphone 1, sound card 2, TIA program 3, microcoder 4, if the device is a physically independent unit 5, and communication medium 6, 7, partly between the device for input of the voice signals 1, and the unit 5/8, and partly between the unit and the device or devices to be controlled 9.

TIA should in the simplest form, Fig. 1, be seen as a "black box" 10, where the voice signal, which represents for example a letter, a number, or other keys and key combinations, comes in and is compared with signals stored in the device, and out will a standard code according to a standardized code system e.g. ANSI, ASCH, ISO, EBEDIC, teletex, digit tones, or the like. It is thus said "a", "b", etc and the device generates, if ASCII code is used, 97, 98, etc, see Fig. 3.

In this way, a logical standard interface 1 1 is created, in order to be able to easily communicate with other equipment 9 which uses the same code system.

The workflow for this can vary in part depending on whether the speech recognition is speaker-dependent or speaker-independent.

In speaker-dependent recognition, for example, the user begins by "training the computer" for the "vocabulary" the computer program should have. Ie. an installer asks the user to speak into the microphone according to the instructions. The user enters the name of some or all of the keys and standard combinations of these, one by one, by entering them.

A sound card digitizes the spoken characters.

The TIA program stores these pronunciations of the names in a register, in which each entry is related to the default code for each of these keys or default key combinations.

After completing the training part, the TIA is ready for use. Now when the user speaks in e.g. a character to the computer, the character is digitized and then ironed with the stored characters in the 502 013 7 TIA program register. When the character is recognized, the default code for the corresponding character is retrieved and forwarded for processing by other programs or devices.

In speaker-independent recognition, the user does not need to train the computer in that way for the current "vocabulary", since the language sounds of these characters, the phonemes, are already stored in the register of the TIA program. Here, a short training of a reference text is enough.

The fact that the sounds of the characters are pre-programmed means that the program works regardless of who the user is, as long as he / she speaks the same (human) language for which the program is pre-programmed.

The TIA program does not need to be visible in any way on the computer screen, just like a regular keyboard driver that is not visible on the screen.

Any error messages can be indicated either on the screen as text, voice or signals.

By configuring the TIA differently for different purposes, such a device can be applied to a computer or other device having a keyboard and / or using a standard code system, to completely replace or complement the keyboard. In this way, control and use of, for example, the computer and all its programs must be able to be performed with only the voice.

Since TIA uses i.a. the alphabet in a language, it has an "unlimited" vocabulary. You can use TIA for writable characters, ie. letters, numbers and other characters, as well as for unwritten characters, ie. control and special characters.

TIA can be used to replace function keys, replace buttons on a control unit, replace commands or special command sequences in a program.

TIA can be used to give aliases to certain characters for special purposes e.g. in computer game contexts, as well as when calling files or custom defined commands.

TIA can be applied to a printer or an electronic typewriter to create a voice-controlled typewriter.

TIA can be applied to a fax machine to create a speechless paperless sending machine.

It can be applied to a mobile phone to create a speech-controlled paperless mobile fax machine for transmitting text messages to e.g. Minicall text or another mobile device. 502 013 8 TIA can be applied to a code lock or authorization control system with personal code.

TIA can be applied to a remote control for voice control of, for example, home electronics with the voice.

TIA can be applied to a telephone or a code lock system for dialing by generating, for example, number tones instead of ASCII code.

TIA also has applications in interactive telecommunications and data communication, voice control of toys, computer games, etc.

TIA shall also be used to completely replace or complement a control unit.

In that case, you control Lex. cursor on a screen by entering the desired position. For example, enter the XY coordinates or the X and Y coordinates, respectively, to move the cursor to the desired position.

TIA has the task of mainly replacing a keyboard and / or a control unit, which can be used e.g. as a disability aid.

This device can be physically constructed as an independent unit 5, with standard physical interfaces 12, 13. Then the output 13 of the unit must fit into the standard input of the equipment to be controlled or used 9.

Communication medium between TIA and other equipment 7, as well as between the voice input and the equipment for voice input 6, can be different, ie. different types of cords or other medium, such as light, modulated radio, etc.

TIA can also be a logic unit 8, i.e. it may be part of, for example, a computer system and use its hardware to control and monitor the system, programs 14, and other equipment connected to the system 9.

The device according to the invention can be used in a number of technical areas, industries and user groups.

TIA can be configured and applied to a range of devices and equipment for voice control of these, or the development of new products. Examples of these are: Voice-controlled computer, voice-controlled typewriter, speech-controlled paperless sending fax machine, voice-controlled paperless mobile fax machine, voice-controlled code lock system or access control with personal code, voice-controlled 502 013 9 remote control, voice-controlled telephone, applications in interactive telecommunications and data communication , speech-controlled computer games etc.

TIA can have many different uses such as disability aids to easily and smoothly control and use a computer, various programs and the equipment, which are connected to a computer. In this way, a disabled person can perform a job or communicate with the outside world. It can be used to improve the working environment, ie. for preventive purposes against occupational injuries due to too much use of computer mouse and keyboard. TIA can be used to improve data security, ie. reduce the risk of incorrect entries at e.g. registration or word processing, to increase productivity, as it is faster to speak than to write, for increased fl flexibility by eliminating the need for keyboards and hands and replacing them with the voice and for example microphone for e.g. inventory management, to reduce chair play on mobile devices by eliminating the need for a keyboard, etc.

The device is of course not limited to the shown, described embodiments, applications and areas of use, but can be varied in your respects within the framework of the idea underlying the device.

Claims (10)

502 015 10 PATENT CLAIMS Device for speech-controlled input of data in the form of codes to an executive unit, such as a computer, a computer program, a code lock system, an electronic typewriter, etc., designed with speech recognition technology, which can be either speaker-dependent or speaker-independent. The device comprises both hardware and software and is characterized in that the device has a logical standard interface for outputting data (11), which device is designed so that the incoming voice signals represent one or a combination of some of all the keys of the keyboard according to the contained in a standard code system is compared and converted in the speech-controlled input device into an output in the form of a standard code or signal according to the standardized code system, which code is then to be transmitted to an executive unit (9, 14) with a standard logic input interface. Device according to claim 1, characterized in that the device is functionally equivalent to a computer keyboard, ie. the device is designed to control menu-driven computer programs and to enter text in standard applications without any adaptation of the applications by using standard code for any or a combination of some of all the keys on the keyboard. Device according to the preceding claims, characterized in that the device has functionally equivalent function keys, designed to be used in applications to perform the functions specific to the particular application, without making any adaptation of the application, by using the standard code. for the function keys and combinations of these with other keys. Device according to any, some or all of the preceding claims, characterized in that the device is designed to control the cursor on a screen by specifying the desired cursor position, for example the XY coordinate or the X and Y coordinates according to a standard code system. 502 013 11 Device according to any, some or all of the preceding claims, characterized in that the device can physically be either an independent unit (5), with a standard physical interface (12, 13), for connection directly to the input of the device. equipment to be controlled, or a logic unit (8), ie. part of another equipment or computer system for controlling these. Device according to any, some or all of the preceding claims, characterized in that the device is applied to a fax machine to create a speech-controlled paperless transmitting fax machine. Device according to any, some or all of the preceding claims, characterized in that the device is applied to a mobile telephone to create a voice-controlled paperless mobile fax machine for transmitting text message to eg Minicall text or another mobile device. Device according to any one, some or all of the preceding claims, characterized in that the device is applied to a code lock or access control system with personal codes. Device according to any one, some or all of the preceding claims, characterized in that the device is applied to a telephone or a code lock system for dialing by generating, for example, numerical tones instead of code according to another standard. Device according to one, some or all of the preceding claims, characterized in that the device has applications in telecommunications and data communication, voice control of toys, computer games, remote control for home electronics, etc.