JPH05307461A - Voice data input/output device - Google Patents

Abstract

PURPOSE:To execute the interactive input of data between a data inputting person and terminal equipment in portable voice data input/output terminal equipment. CONSTITUTION:The voice data is inputted from a microphone 18. The inputted voice data is recognized by a voice recognition part 11, and is analyzed by a CPU 12. In the case that there is inconsistency in the data or there is inquiry data as the result of the analysis of the voice data, the CPU 12 supplies the voice data for interaction to a voice synthesis part 15 in order to answer it. Then, a synthetic voice for the interaction is outputted from a speaker 17. An operator inputs the voice data from the microphone 18 in order to answer that synthetic voice for the interaction. Thus, even if the operator is inexperienced, he can input correct data successively without being puzzled at operation.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice data input / output device which can be used for data collection, work instruction, data search and the like. In particular, the present invention relates to a portable voice data input / output terminal device.

[0002]

2. Description of the Related Art Conventionally, there is known a portable voice data input / output terminal device which allows an operator to input data by voice. For example, JP-A-59-27346 and JP-A-61-290530 disclose this type of portable voice data input / output terminal device.

According to the portable voice data input / output terminal device described in Japanese Patent Laid-Open No. 59-27346, the measurement data read by each measuring instrument from each instrument is input as voice to the portable radio device, and this voice is transmitted. The data is supplied to the voice recognition device via the transmission line. Then, the voice data is recognized by the voice recognition device, and the recognized voice data, that is, the measurement data is stored as digital data. In addition, the recognized measurement data is returned to the portable wireless device via the transmission line and is appropriately confirmed by the measurement staff.

According to the one disclosed in Japanese Patent Laid-Open No. 61-290530, necessary items are provided in accordance with a voice output command from a host computer via a helmet-equipped wireless voice input device while a worker is landing. Input voice. Then, the input information is fed back and voice-outputted to the input person so that the operator can check the input information. The input information is checked by inputting “yes” if the feedback voice output is exactly as input by the worker, and inputting “no” if it is incorrect.

A similar data input method is disclosed in, for example, Japanese Unexamined Patent Publication No. 3-15099 as well as the above-mentioned publication.

[0006]

According to the conventional portable voice data input / output terminal device capable of inputting / outputting data by voice, a worker or the like can input data while engaging in work. Moreover, since the input data can be confirmed by the synthesized voice, there is an advantage that the accurate data can be input while continuing the predetermined work.

However, in the conventional voice data input / output device, as described in, for example, Japanese Patent Application Laid-Open No. 61-290530, data input / output is performed by a predetermined procedure defined by a management computer or the like. However, there is a drawback that the data input / output procedure is poor in flexibility. Therefore, in the case of the conventional voice data input / output device, it was not possible for the operator to ask a question and to answer the question with respect to the input instruction content output by the synthesized voice. On the contrary, when the input voice of the operator cannot be recognized, it is not possible for the device side to make an inquiry to the operator. That is, according to the conventional voice data input / output device, although the voice data input by the operator can be confirmed, the voice data can be input only by a predetermined procedure. Therefore, even if the operator does not understand the meaning of the voice instruction given from the device during the operation, the operator must continuously input the voice data, and the certainty of the voice data input cannot be sufficiently guaranteed. was there.

Even if the voice data input / output device cannot correctly understand the meaning of the input data input by the operator, if the operator confirms that the input voice data is correct, the device cannot input the correct meaning. The processing had to be carried out based on the data, and there was a possibility of performing incorrect processing. Therefore, according to the present invention, a data input / output device is provided with an interactive function with an operator, and data input / output between the operator and the device is interactively performed, whereby data input / output operation by voice can be efficiently performed. An object is to provide an audio data input / output device.

[0009]

A voice data input / output device according to the claims is a voice recognition means for recognizing input voice data, and an analysis for analyzing the type of voice data recognized by the voice recognition means. Means, a dialogue control means for deciding the data to be output based on the analysis result of the analysis means, and a synthetic voice output means for outputting the data decided by the dialogue control means as a synthetic voice. is there.

[0010]

The operation of the present invention will be described with reference to FIG. The input voice is given to the voice recognition means 1. The voice recognition means 1 is provided with a dictionary registered in advance. The voice recognition unit 1 recognizes what the input voice has by referring to the dictionary. Therefore, the operator can input data by voice and does not need to operate a keyboard or the like.

The content recognized by the voice recognition means 1, that is, the recognition data, is given to the analysis means 2. Analysis means 2
Analyzes the recognition data and determines whether or not there is a contradiction or an error in the data type or data. The analysis result is given to the dialogue control means 3 together with the recognition data.
The dialogue control means 3 determines the contents to be interacted with the operator who inputs the voice data based on the recognition data and the analysis result. The content to be interacted with is, for example, to determine a voice for prompting re-input or requesting correction of the input when there is a contradiction or an error in the input voice. For ambiguous input data, a voice for inquiring about an ambiguous point is determined. Then, the decided contents are given to the voice synthesizing means 4.

The voice synthesizing means 4 converts a character string expressing a voice into an actual voice and outputs it as a synthesized voice. Therefore, the operator does not need to look at the display or the like to receive the output data, and can confirm the data content by hearing rather than visually. The operator, who hears the synthesized voice output from the voice synthesizing means 4, inputs the corresponding voice to the apparatus again.

[0013]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 2 is a block diagram showing the configuration of a portable voice data input / output terminal device according to an embodiment of the present invention. The terminal device 10 includes a voice recognition unit 11 for recognizing an input voice, a CPU 12 and a CPU 12 for processing a recognition result recognized by the voice recognition unit 11.
ROM 13 in which the operating program and dictionary of
R as a memory utilized when the CPU 12 processes
An AM 14, a voice synthesizing unit 15 for converting output data into synthetic voice, and a battery 16 for supplying driving power to the input / output terminal device 10 are provided. Further, a headset having a speaker 17 and a microphone 18 is attached to the portable voice data input / output terminal device 10 as a detachable external device.

The above-described voice recognition unit 11 is the microphone 1
It is for recognizing voice data by converting an analog voice signal inputted from 8 into a digital signal. CP
The processing in U12 includes analysis processing and dialogue control processing. In the analysis processing, the type of the recognized voice data given from the voice recognition unit 11 is analyzed, and it is determined whether or not there is a contradiction or an error in the content. In the dialogue control process, the character string data expressing the synthetic speech given to the speech synthesis unit 15 is determined.

The voice synthesizing unit 15 converts the character string data sent from the CPU 12 into an analog signal which is a synthetic voice and outputs it to the speaker 17. FIG. 3 is a flowchart showing an example of processing of the CPU 12 in FIG. Next, a description will be given along the flow of FIG. 3 with reference to FIG. The following description will be made, for example, when the worker of the electric power company possesses the portable voice data input / output terminal device 10 and inspects whether the transformer provided on the electric pole is operating normally. The case of using for result input work will be described as an example.

When the operation switch (not shown) of the portable voice data input / output terminal device 10 is turned on and the operation is started, the character string data representing the voice to be synthesized is given from the CPU 12 to the voice synthesizing unit 15. The speech synthesis unit 15 generates synthetic speech. Then, the sound is output from the speaker 17 of the headset (step S1). For example, a synthetic voice "Please input the surveyed device number" is output.

The CPU 12 determines whether or not the recognition data from the voice recognition unit 11 is present (step S2). If the recognition data is not present within a predetermined time, the voice synthesis unit 15 is operated.
Again, give the character string data expressing the voice to be synthesized,
The voice synthesizer 15 outputs the same synthesized voice. This is repeated while there is no recognition data, and the state of waiting for recognition data is continued.

The operator inputs data by voice using the microphone 18 according to the synthesized voice output from the speaker 17. For example, "Transformer C of utility pole K1
Enter the number of the inspected part, such as "3". The voice data input from the microphone 18 is stored in the voice recognition unit 1
1 to convert the analog voice signal into a digital signal to recognize the content of the voice data. The recognized data is the CPU 12 in the recognition data waiting state.
Sent to.

When it is determined in step S2 that the recognition data is present, the recognition data is sent to the voice synthesizing unit 15 (step S3), the voice synthesizing unit 15 generates a synthetic voice, and the synthetic voice is generated by the speaker 17. Is output (step S4). Therefore, the operator can confirm the voice data that he / she has input from the microphone 18 by the synthesized voice output from the voice synthesizer 15 via the speaker 17. And at the time of this confirmation, if the synthesized voice matches the voice data input by itself,
The operator inputs “yes” to the microphone 18. On the other hand, if the voice input by itself and the synthesized voice that is output are different, it means that the voice recognition unit 11 has not correctly recognized the input voice, so "No" is input.

The entered "Yes" or "No" is
It is recognized by the voice recognition unit 11 and given to the CPU 12.
The CPU 12 confirms whether the recognition data is "yes" (step S5). As a result, if it is confirmed to be “Yes”, the input data previously input (step S
The data recognized in 2) is analyzed (step S).
6).

On the other hand, if the recognition data is not "yes", the inquiry voice is determined (step S1).
0). For example, "Please enter the data again."
The inquiry voice is determined. Then, this is converted into a synthetic voice by the voice synthesizing unit 15 and output from the speaker 17. In response, the operator will input the voice data again.

Data analysis by the CPU 12 (step S
After 6), it is determined whether or not the data is a proper type of data, and whether or not the data has a contradiction or an error (step S).
7). If the data is proper, that data is adopted (step S8). The adopted data is stored in the RAM 14. Then, the character string data representing the content of the instruction is determined by the CPU 12 so that the operator next designates the data to be input (step S9). Then, the data is given to the voice synthesizing unit 15 to be a synthesized voice and output from the speaker 17 (step S1). For example, this synthetic voice is "Please enter the presence or absence of failure."
Etc. The operator follows the instruction and the microphone 1
8 will be used to enter the next data.

When it is determined in step S7 that the input data is not proper, the CPU 12 determines inquiry data for dialogue (step S10). For example, "Please enter the data again." Or "The data you have just entered is inconsistent with the data you entered earlier.
Please re-enter the correct data. Inquiry data such as “” is determined. Then, the inquiry data is converted into a synthetic voice by the voice synthesizing unit 15 and output from the speaker 17 (step S1).

If the data adopted in step S8 is inquiry data from the operator, for example, "what kind of content is the current instruction?", The next data determined in step S9 is the operator. Will be the data to answer the inquiry from. As described above, according to this embodiment, since the operator and the portable voice data input / output terminal device 10 can interactively exchange data and interactively input data, correct data can be accurately obtained. You can enter. In particular, even an operator unfamiliar with data input can easily perform correct data input.

In the above structure, a wireless section is provided in the portable voice data input / output terminal device 10, and wireless communication can be performed between the CPU 12 and the host computer arranged at a predetermined location through the wireless section. By doing so, CPU1
The host computer can also share the dialogue control processing and analysis processing performed by 2, and more complex dialogue is possible. Further, since the data can be stored in the large-capacity storage device of the host computer, the RAM 14 in the above configuration can be downsized and the configuration of the portable voice data input / output terminal device 10 can be downsized.

In the above embodiments, for example, a portable voice data input / output terminal device for a worker who climbs on a utility pole and works is taken as an example. However, the present invention is an operation program provided in the ROM 13. It is possible to make the input / output device suitable for various uses by simply changing the. Further, although the above embodiment has been described by taking the portable voice data input / output terminal device as an example, the present invention can be applied to a voice data input / output device which is not portable.

In addition, the present invention can be variously modified in accordance with the configurations described in the claims.

[0028]

EFFECTS OF THE INVENTION Since the present invention is configured as described above, data input is performed interactively between the operator and the apparatus, and if the meaning is unknown, it can be confirmed and correct. You can enter data. Further, the operator does not need to use his / her hands for data input, and can correctly input data without checking the input data on the display or the like. Therefore, the efficiency of data input work can be improved.

[Brief description of drawings]

FIG. 1 is a diagram for explaining the operation of the present invention.

FIG. 2 is a configuration block diagram of a portable voice data input / output terminal device according to an embodiment of the present invention.

3 is a flowchart for explaining the operation of the voice data input / output terminal device shown in FIG.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 voice recognition means 2 analysis means 3 dialogue control means 4 voice synthesis means 10 portable data input / output terminal device 11 voice recognition section 12 CPU 15 voice synthesis section

Claims (1)

[Claims] 1. A voice recognition means for recognizing input voice data, an analysis means for analyzing a type of voice data recognized by the voice recognition means, and a data to be output is determined based on an analysis result of the analysis means. A voice data input / output device, comprising: a dialogue control means for performing the above; and a synthesized voice output means for outputting the data determined by the dialogue control means as a synthesized voice.