JPH0883097A - Device operation instruction method and device for excuting the method - Google Patents

Abstract

PURPOSE: To improve the efficiency of the human interface by referring to the past synthesized voice instruction outputs while outputting synthesized voice instructions based on the recognition results of human motion/operations and machine operations. CONSTITUTION: In the case of instruction on the operations of putting a car into a garage, for example, a distance computing section 201 receives the data obtained by a sensor and computes the distance between the body of the car and the wall of the garage. An instruction voice storage memories 202 to 205 store the instruction voices synthesized and outputted in the past or the contents of the voices. A condition judgement processing section 206 makes a judgement on the condition based on the distance between the body and the wall computed by the section 201 and the past instruction voice outputs and decides to generate a proper response sentence. These synthesized voice response sentences are beforehand registered in a response voice storage device 207. A response sentence selection processing section 208 selects a synthesized response sentence corresponding to the judgement processing result of the section 206 from the synthesized voice response sentences stored in the section 207 and outputs the selected sentence.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a device operation instructing method and an apparatus for implementing the method, and more particularly to recognizing human motions and / or operations and device motions, and appropriately synthesizing them based on the recognition result. The present invention relates to a device operation instruction method for improving the efficiency of a human interface by outputting a voice, and an apparatus for implementing this method.

[0002]

2. Description of the Related Art As the downsizing of computers progresses, the chances of using the computers increase, and it is not uncommon to use the computers for form preparation work, document preparation and other daily work. In addition, the number of cases where beginners unfamiliar with computers use it is increasing. Considering such a situation, the human interface of a computer becomes important.

[0003] For example, when the form creation is a daily work, there are cases in which the form entry field is mistaken due to an inadvertent mistake due to familiarity with the form creating work or due to tiredness of continuous work. In such a case, by pointing out an error and instructing how to deal with it by voice, the error can be easily recovered and the waste of continuing the erroneous work can be omitted. Although the error can be visually pointed out, the eyes are always paying attention to the data to be input, and it is considered that it is more effective to rely on the voice to immediately notice the error. Also, in order to give various instructions to a beginner unfamiliar with a computer, it may be better to rely on voice rather than on vision. For example, since various kinds of information are displayed on the screen of a computer, a beginner has a hard time even finding where on the screen the information he or she needs is displayed. In such a case, a beginner can learn the operation of the computer smoothly by giving voice instructions at a timely timing.

The instruction by the voice output is applied not only to the interaction with the computer equipment but also to the cooperative work between the human and the equipment in general, or to monitor the human operation and give attention to the operation. But it is effective. For example, when the driver reverses the car and puts it in the garage, the distance between the car and the garage wall is measured, and the steering wheel operation is "turned to the right" or "after" according to the distance. You can move back a few centimeters. " It is known from the daily experience that such an operation instruction by voice is very convenient.

[0005]

As described above, in order to improve the efficiency of the human interface between the equipment and the person who operates the equipment, the instruction itself using the voice output has been performed conventionally. There is. However, it is hard to say that the conventional example of the device operation instructing apparatus that uses voice output outputs only a message registered in advance and outputs appropriate voice according to various situations. It is not possible to expect much improvement in the efficiency of the human interface by simply outputting the registered message without making a situation judgment.

The present invention recognizes a human motion and / or an operation and a device motion, and outputs an appropriate voice based on the recognition result to improve the efficiency of a human interface and a method for this device operation. An apparatus for carrying out the method is provided.

[0007]

In a device operation instruction method for recognizing an operation and / or an operation and an operation of a device and generating a synthetic voice instruction output based on a recognition result, a past synthetic voice instruction output is referred to. A device operation instruction method for generating a synthetic voice instruction output is configured. And action and /
Alternatively, a detection device 101 for detecting the operation and the operation of the device,
A synthetic voice response that includes 102 and includes instruction output memories 202 to 205 that store past synthetic voice instruction outputs, and that generates a synthetic voice instruction output based on the detection result of the detection device and the stored contents of the instruction output memory. Sentence generator 20
A device operation instruction device including 7, 208, and 209 was configured.

Further, the synthesized voice response sentence generating device constitutes an equipment operation instructing device which is an edited voice synthesizing device or a rule-based voice synthesizing device. Further, the synthesized voice response sentence generation device constitutes a device operation instruction device having a voice conversion processing unit 209 for changing the shape of the fundamental frequency pattern of the voice and / or the duration of the phoneme.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT An embodiment of the present invention will be described with reference to FIGS. FIG. 1 is a diagram showing the overall configuration of a device operation instruction device of the present invention. FIG. 2 is a diagram showing a synthetic voice response sentence generation device which constitutes the device operation instruction device. In FIG. 1, 101 is a keyboard, mouse, touch panel, and other input devices, and 102 is a magnetic sensor, an optical sensor, an image recognition device, a voice recognition device, and other sensors that detect physical quantities. With these input devices and sensors, data necessary for recognizing human motion, human operation, and device motion can be obtained. An output device 103 outputs a voice and / or an image, and outputs an instruction voice or displays the state of the device body using these. Reference numeral 104 denotes a main unit, which includes a storage device indicated by 105 and an arithmetic processing unit indicated by 106. The main device 104 recognizes a human motion, a human operation and a device motion based on the data detected by the input device 101 and the sensor 102, and generates a response sentence. A disk device 107 is a storage device that stores data and programs used in the main device 104.

Here, the method of instructing the device operation will be specifically described by taking a case of a car garage as an example. Referring to FIG. 2, reference numeral 201 denotes a distance calculation unit that inputs data obtained by the sensor 102 and calculates the distance between the vehicle body of the automobile and the garage wall. Reference numerals 202 to 205 denote instruction voice storage memories for storing the instruction voice or the voice content of the synthesized voice output in the past. This example shows 4
Each instruction voice is stored. A situation determination processing unit 206 determines the situation based on the distance between the vehicle body and the wall calculated by the distance calculation unit 201 and the past instruction voice output, and determines what kind of response sentence should be generated. For example, a synthetic voice response sentence such as "10 cm to the rear wall" and "turn the steering wheel a little more to the right" is generated. Situation determination using past instruction voice output means "cut to the left" even though the voice instruction "cut to the right" was given immediately before.
If it is, "right, right" is instructed, or if there is little "cutting", "cut more" is referred to, and an instruction to further emphasize this is given by referring to the previous voice instruction. It is assumed that such a synthetic voice response sentence is registered in the response voice storage device 207 in advance. It is also possible to use a rule-based speech synthesizer without using registered voice. A response sentence selection processing unit 208 selects and outputs an appropriate synthesized response sentence corresponding to the determination processing result of the situation determination processing unit 206 from the synthesized voice response sentence stored in the response voice storage device 207. Reference numeral 209 denotes a voice conversion processing unit, which performs a process of changing the fundamental frequency and duration of the voice to increase variations of the response sentence. When "right, right" is instructed, it is possible to make the expression tense by increasing the fundamental frequency and shortening the duration.

[0011]

As described above, the present invention recognizes a human motion, a human operation, and a device motion, and provides a device operation instruction by an appropriate synthesized voice output based on the recognition result. Therefore, the present invention has an effect of facilitating operation of the device, improving work efficiency, and lightening human stress caused by operation of the device.

[Brief description of drawings]

FIG. 1 is a diagram showing an example.

FIG. 2 is a diagram showing a synthetic voice response sentence generation device.

[Explanation of symbols]

 101 input device 102 sensor 103 output device 104 main device 105 storage device 106 arithmetic processing device 107 disk device 201 distance calculation unit 202 to 205 instruction voice storage memory 206 situation determination processing unit 207 response voice storage device 208 response sentence selection processing unit 209 Voice conversion processing unit

Claims (4)

[Claims] 1. A device operation instruction method for recognizing an operation and / or an operation and an operation of a device and generating a synthetic voice instruction output based on the recognition result, in a synthetic voice instruction by referring to a past synthetic voice instruction output. A device operation instructing method characterized by generating an output. 2. A detection device for detecting an operation and / or an operation and an operation of a device, and an instruction output memory for storing a past synthetic voice instruction output, and a detection result of the detection device and an instruction output memory. An apparatus operation instruction device comprising a synthetic voice response sentence generation device for generating a synthetic voice instruction output based on stored contents. 3. The device operation instruction device according to claim 2, wherein the synthesized voice response sentence generation device is an edited voice synthesis device or a rule-based voice synthesis device. 4. The device operation instruction device according to claim 3, wherein the synthetic voice response sentence generation device has a voice conversion processing unit for changing the shape of the fundamental frequency pattern of the voice and / or the duration of the phoneme. An apparatus operation instruction device characterized in that