JPH0215897B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a voice information input device that recognizes information input by a human voice and processes the information appropriately.

[Conventional technology]

In the case of collecting data, such as inventory taking at a store or memorizing readings of measuring instruments at various plants, when the amount of data (types of products in inventory or number of measuring instruments at a plant) is large, the person collecting the data is Checking and recording each item takes a lot of time and effort. Therefore, an apparatus for collecting data by voice has been proposed, for example, in Japanese Patent Application Laid-Open No. 169896/1983.

[Problem to be solved by the invention]

By the way, in general, in data acquisition as described above, the surroundings of the acquisition location are rarely quiet, and there are many cases where a considerable amount of noise is generated. Therefore, there is a risk that the above-mentioned device may mistakenly recognize ambient noise as voice input and capture it as data. Currently, in conventional devices, no effective countermeasures are taken against such noise.

The purpose of the present invention is to solve the above-mentioned conventional problems,
To provide a voice information input device capable of improving the speech recognition rate even when noise is generated around a speaker.

[Means to solve the problem]

In order to achieve the above object, the present invention provides a microphone that converts a speaker's voice into an electrical signal, a voice recognition device that inputs the electrical signal from the microphone and recognizes the speaker's voice, and A voice information input device comprising a voice control device that performs predetermined processing based on the voice of the speaker recognized by the voice recognition device, wherein the voice information input device includes a voice control device that performs predetermined processing based on the voice of the speaker recognized by the voice recognition device. The present invention is characterized by being provided with a noise signal generator that generates an audio signal that simulates noise.

Still another invention is characterized in that, in addition to the above-mentioned invention, a speaker disposed around the speaker and an audio signal output means for outputting an audio signal to the speaker are provided.

[Effect]

The voice of a particular speaker and the characteristics of the simulated noise around the speaker produced by the noise signal generator are stored in the speech recognition device. At the work site, when the speaker inputs voice through a microphone, surrounding noise is also input at the same time, and this is input to the voice recognition device, which recognizes only the voice of the speaker. A predetermined process is performed based on.

Further, in the other invention, in addition to the above effect, a predetermined sound is output from speakers around the speaker.

[Example] The present invention will be described below based on the illustrated example.

FIG. 1 is a system configuration diagram of a voice information input device according to an embodiment of the present invention. In the figure, 1 is a voice information input device, 2 is a radio fixed station, and 3 is a radio mobile station carried by a speaker who is a data collector. 4
is a microphone, and 5 is an earphone speaker. The radio fixed station 2 and the radio mobile station 3 are coupled by radio waves.

The voice information input device 1 includes a voice recognition device 11, a voice output device 12, a voice control device 13, a noise signal generator 14, and an acoustic signal generator 15. As is well known, the voice recognition device 11 is a device that registers a specific voice in advance, memorizes its characteristics, and recognizes the data of the specific voice from input received voice data, and the voice output device 1.
2 is a device that generates a guidance sound (a voice that instructs input or repeats input data) for the speaker; a voice control device 13 determines the voice recognition result, takes in the data, and generates the guidance sound;
This is a device that performs various other processes.

The noise signal generator 14 is a device that generates a signal corresponding to a simulated noise similar to the surrounding noise of the speaker. For example, an audio frequency transmitter is used, and the audio signal of the noise signal generator 14 is used for speech recognition. It is registered in the device 11 and its characteristics are stored. Acoustic signal generator 1
5 is a device that generates a signal corresponding to some kind of noise or music. The noise signal generator 14 and the acoustic signal generator 15 are provided with a volume adjuster (not shown), the output level of which is adjusted in advance according to the intensity of the surrounding noise of the speaker.

Next, the operation of this embodiment will be explained. A speaker who inputs voice information carries a radio mobile station 3 and inputs voice data through a microphone 4 by speaking. At this time, since sound (noise or music) is always being played from the earphone speaker 5 in accordance with the signal output from the audio signal generator 15,
Regardless of the presence or absence of surrounding noise, speakers unconsciously speak relatively loudly because they feel that there is surrounding noise.
In addition, voice data is always input in a constant tone.

In this way, the signal level of the audio data input to the microphone 4 has a higher level than when the above-mentioned sound is not played. On the other hand, a noise signal similar to the surrounding noise of the speaker is input from the noise signal generator 14 to the speech recognition device 11, and the speech recognition device 11 recognizes this. Therefore, even if a surrounding noise signal is input together with the voice data input signal from the microphone 4, the voice recognition device 11 can identify this and
It is not recognized as voice. This, combined with the high level of the voice data input signal, eliminates the noise surrounding the speaker and ensures that the voice data is recognized. Input of audio data can be confirmed by listening to the answer back from the audio output device 12 through the earphone speaker 5.

In this way, in this example, the signal from the noise signal generator is input to the speech recognition device and recognized, and the surrounding noise of the speaker is not recognized as speech, so that the speech data can be reliably recognized. can do. In addition, since the acoustic signal generator always generates sound, the speaker unconsciously inputs audio data out loud, which increases the signal level. Data can be reliably recognized. Also, the sound from the acoustic signal generator is made into music, and this music is digitized and recorded on an integrated circuit, and the A/D
If the music is converted by a converter and played through an earphone speaker, it is possible to input audio data without causing discomfort to the speaker.

FIG. 2 is a system configuration diagram of a voice information input device according to another embodiment of the present invention. In the figure, parts that are the same as those shown in FIG. This embodiment differs from the previous embodiment in that voice data is input to the voice information input device 1 via the radio fixed station 2 and the radio mobile station 3, whereas the present embodiment This embodiment differs only in that audio data is directly input to the audio information input device 1, and its operation and effects are the same as those of the first embodiment.

FIG. 3 is a system configuration diagram of a voice information input device according to still another embodiment of the present invention. In the figure, parts that are the same as those shown in FIGS. 1 and 2 are designated by the same reference numerals, and description thereof will be omitted. This embodiment differs from the previous embodiments in that while the previous embodiments were equipped with an acoustic signal generator 15, this embodiment is omitted. In this embodiment, the acoustic signal generator 1 of each of the previous embodiments is
5, the noise signal generator 14 is also used. The signal from the noise signal generator 14 is output to the earphone speaker 5, and since noise is always generated from the earphone speaker 5, the speaker perceives this as noise regardless of the presence or absence of surrounding noise, and naturally The user can now speak loudly, and therefore, a clear voice that can overcome the noise can be input to the voice recognition device 11. The other operations of this embodiment are the same as those of the previous embodiments. Further, the effects of this embodiment are the same as those of the previous embodiments, and in addition, since the acoustic signal generator can be omitted, the configuration of the apparatus can be simplified.

In the description of the above embodiment, an example was described in which both a noise signal generator and an acoustic signal generator are used, or an acoustic signal generator is used as a noise signal generator. If training is being performed or if the noise at the work site is relatively low, it is not necessarily necessary to use the audio signal generator itself or the noise signal generator.
Further, the earphone speaker can be a normal speaker. Furthermore, it goes without saying that the audio information input device of the present invention is applicable to purposes other than data acquisition. Furthermore, although an example has been described in which signals are output to the voice recognition device and the speaker at the same time, it is also possible to select only one of them.

〔Effect of the invention〕

As described above, in the present invention, a noise signal generator is provided so that a signal simulating the noise around the speaker can be input to the speech recognition device, so that the speaker's voice signal is input to the speech recognition device. In this case, it is possible to prevent noise signals around the speaker from being recognized as speech, thereby making it possible to reliably recognize speech data and improve the recognition rate.

Further, in another invention, by outputting sound from a speaker, the sound level of the speaker can be increased without conscious effort from the speaker, and as a result, the reliable recognition of sound data and its recognition rate are improved. can be further improved.

[Brief explanation of drawings]

FIGS. 1, 2, and 3 are system configuration diagrams of audio information input devices according to embodiments of the present invention, respectively. 1... Voice information input device, 4... Microphone, 5... Earphone speaker, 11... Voice recognition device, 12... Voice output device, 13... Voice control device, 14... Noise signal generator, 15 ...Acoustic signal generator.

Claims (1)

[Scope of Claims] 1. A microphone that converts a speaker's voice into an electrical signal; a voice recognition device that receives the electrical signal from the microphone and recognizes the speaker's voice;
A voice information input device that is connected to an input end of the voice recognition device and that performs predetermined processing based on the voice of the speaker recognized by the voice recognition device. 1. An audio information input device comprising a noise signal generator that generates an audio signal that simulates noise. 2. A microphone that converts a speaker's voice into an electrical signal; a voice recognition device that receives the electrical signal from the microphone and recognizes the speaker's voice;
A voice information input device that is connected to an input end of the voice recognition device and that performs predetermined processing based on the voice of the speaker recognized by the voice recognition device. Audio information comprising a noise signal generator that generates an audio signal simulating noise, a speaker placed around the speaker, and audio signal output means that outputs an audio signal to the speaker. input device. 3. In claim 2, the audio signal output means includes an acoustic signal generator that generates an electrical signal corresponding to a predetermined sound, and a signal transmission means that transmits the electrical signal of the acoustic signal generator to the speaker. A voice information input device comprising: 4. The audio information input device according to claim 2, wherein the audio signal output means is a signal transmission means for transmitting an electrical signal from the noise signal generator to the speaker.