JP3082742U - Microphone for speech recognition system input - Google Patents

Abstract

(57) [Summary] To provide a voice input means suitable for use in a voice recognition system or the like in an electronic medical record. A microphone for an input of a voice recognition system provided with a function similar to a receiver of a conventional telephone, that is, an earpiece, a voice input part, and a handle part, wherein the earpiece part and the voice input part are a handle. The earpiece and the voice input unit are provided at the handle so that the voice input unit is located near the human mouth at the same time that the earpiece is put on the human ear. It has been found that the above problem can be solved by providing a microphone for inputting a speech recognition system.

Description

[Detailed description of the invention]

[0001]

[Technical field to which the invention belongs]

 The present invention relates to a voice recognition system input microphone to be used in a computer voice recognition system. More specifically, the microphone for input of the voice recognition system according to the present invention is used to record the patient's chief complaint, medical history, findings, etc. in an electronic medical record system (system for recording medical records electronically) used by doctors during medical treatment. This is a device relating to a microphone for inputting a speech recognition system suitable for use when a speech recognition system is used at the time of input.

[0002]

[Prior art]

 Conventionally, most microphones (microphones) used in a voice recognition system are a headset type microphone 10 (FIG. 1) fixed to a head 12 by a hair band 11, and only a small part of a fixed microphone 20 (second microphone) is used. FIG.) And hand microphone 30 (FIG. 3) are used. This is because voice recognition systems are generally vulnerable to noise and changes in sound loudness (level), and if the positional relationship (distance or angle) between the speaker's mouth and microphone is not constant, the voice recognition rate drops significantly. This is because

[0003] In other words, in order to eliminate ambient noise as much as possible, a microphone having a fairly strong unidirectionality is used. If the microphone is not correctly directed to the sound source (the mouth of the speaker), the sound of the voice will be lost. Although the recognition level is extremely lowered, in the case of the headset type microphone 10, the microphone 13 is fixed to the mouth of the speaker by the hair band 11 and the beam 14, so that the speaker moves. Regardless of the posture or posture, the positional relationship between the speaker's mouth and the microphone does not change, and voice can always be input to the microphone at a stable level. On the other hand, in the fixed microphone 20 (FIG. 2), the positional relationship between the microphone and the sound source changes when the speaker moves slightly or the face is tilted, and the level greatly changes. The hand microphone 30 (FIG. 3) can follow the movement of the speaker to some extent, but it is difficult to keep the microphone in a fixed position, and the position of the microphone lowers without being noticed due to fatigue of the arm muscles. Often there is.

[0004] For these reasons, the use of a headset microphone in a speech recognition system has become almost common sense and is becoming established.

[0005]

[Issues solved by the invention]

 When using a speech recognition system at home or office, there is no problem using a headset microphone, but when using a speech recognition system in an electronic medical record, problems unique to the electronic medical record are recognized. .

First, it is assumed that a doctor who is a general user of an electronic medical record wears a headset-type microphone during medical treatment, so that the patient may feel uncomfortable. It is difficult to make an inquiry and palpation with the microphone attached.

[0007] Furthermore, doctors often walk around examination rooms and treatment rooms for examinations and treatments during examinations, and often use a stethoscope or the like. In this case, disconnect the microphone cable from the computer. Or, it is necessary to remove the headset microphone, which is very complicated. Also, when removing and reattaching the headset microphone, it takes time to carefully re-adjust the position of the microphone, etc., and in Japanese medical treatment, which is said to be 3 minutes of medical care per patient. Such an operation is not practical.

[0008] Furthermore, doctors are often unfamiliar with speaking into empty spaces or computers, and the use of a headset-type microphone in electronic medical records makes the speaker uncomfortable. However, the utterance became unnatural or the voice was inevitably reduced.

[0009] Although the fixed microphone and the hand microphone have less discomfort to the patient than the headset microphone, there is a problem that the volume (level) becomes unstable as described above, and the recognition accuracy is low. There is a major drawback of lowering. Also, although it is not as good as a headset microphone, it is undeniable that a doctor talking to the microphone during medical treatment looks strange from the patient's perspective.

Therefore, a problem to be solved by the present invention is to provide voice input means suitable for use in a voice recognition system or the like in an electronic medical record.

[0011]

[Means for Solving the Problems]

 In order to solve the above-mentioned problems, the present inventor studied the voice input mode of the voice recognition system. As a result, they found that the above-mentioned problem could be solved by making the microphone for speech recognition system input into a shape similar to a telephone handset, and completed the present invention.

That is, the present invention comprises an earpiece, a voice input part, and a handle part. The earpiece part and the voice input part are connected by the handle part, and the earpiece part is connected to the human ear. At the same time, the ear input section and the voice input section are provided in the handle section so that the voice input section is located near the mouth of the human, and the microphone for inputting a voice recognition system (hereinafter also referred to as the present microphone). It is a device that provides

According to the present invention, “the earpiece, the voice input part, and the handle part are connected to each other by the handle part, and the earpiece is put on the human ear. At the same time, the earpiece and the voice input unit are provided in the handle so that the voice input unit is located near the human mouth. '' This indicates that the shape conforms. In other words, the user of the microphone generally grasps and lifts the handle of the microphone, places the earpiece on one of the left and right ears, and simultaneously brings the voice input unit close to the mouth. To the voice input unit, the content to be recognized by the voice recognition system can be input by speaking verbally.

The subject to which the microphone can be used is not particularly limited, and can be widely used in any case where an input is made to a speech recognition system. Specifically, it can be used for general voice input to electronic medical records, reservation / reception systems, translation systems, word processors, and other computers, but the features of the present invention are most effectively demonstrated. One of the objects that can be done is electronic medical records.

That is, the present invention relates to a microphone for voice recognition which does not give a patient a sense of incongruity and can stabilize the sound volume (level) even when a doctor is in medical care. Has the meanings listed below.

Earpiece, voice input, and handle 1. Stabilization of sound volume (level) Most people have been accustomed to telephone since childhood, and when they picked up the handle corresponding to the handset, they pushed the earpiece almost reflexively into the ear, The voice input is positioned at an appropriate position near the mouth. This is inevitably done in order to put one's own voice into the microphone while listening to the other party's voice over the telephone, but as a result the ear is fixed and the handset is semi-fixed to the head. The positional relationship between the mouth of the speaker, which is the sound source, and the microphone is substantially constant. Therefore, although the microphone is held by hand, it is the same as the hand microphone, but the position does not become unstable unlike the hand microphone.

[0017] 2. The patient does not feel uncomfortable Unlike the headset microphone, the handset is a very common device, and even if the doctor speaks to it appropriately, there is almost no discomfort on the viewer's side.

[0018] 3. Unlike a headset microphone, a fixed microphone, or a handheld microphone that cannot be used by a doctor who is a speaker, speaking to the receiver has almost no discomfort and the same volume of voice as when calling Allows natural voice input.

[0019] 4. It doesn't have to be put on and taken off. Unlike a headset microphone, the act of putting on and taking off is not required. In the case of a headset microphone, it takes time to put a band on the head with both hands, adjust the beam, and place the microphone in an appropriate position before using the microphone. It takes time to remove the hair from your head and fix your hair. On the other hand, in the case of a handset microphone, when you want to input voice, you just need to pick it up and speak it in exactly the same way as a telephone.When you use a stethoscope or end voice input, you can use the same way as a telephone All you have to do is hang up the handset. It does not take time and effort and does not disturb the medical treatment. It is a common practice for doctors to call during a consultation, and voice input is possible by extending daily activities.

In the present invention, what is particularly important is the shape of the microphone. First, the earpiece and the voice input part are connected by the handle part, and at the same time that the earpiece is brought into contact with the human ear, The earpiece and the voice input unit need to be provided in the handle so that the voice input unit is located near the mouth of the human. In addition, it is similar to a conventional handset, so that it is difficult for the surroundings to feel uncomfortable.When the earpiece is placed on the ear, the voice input unit is positioned at an appropriate position near the speaker's mouth. It is preferable that the length of the handle portion, the angle of the ear contact portion and the voice input portion with respect to the handle portion, the structure of each of these, and the like are appropriately designed ergonomically.

[0021]

[Embodiment of the invention]

 Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 4 is a schematic diagram showing one embodiment of the microphone. In FIG. 4, the earpiece 41, the voice input part 42, and the handle part 43 are integrally molded to constitute the microphone 40. The earpiece 41 and the voice input part 42 have a shape protruding from the handle part 43 in substantially the same direction. When the earpiece 41 is brought into contact with the human ear, the voice input part 42 is placed near the human mouth. It is located. From the back of the voice input unit 42, an input code 44 for inputting a voice signal input from the voice input unit 42 to a computer system in which software to be voice input such as an electronic medical record is installed is provided. And a voice input device (not shown) which is a main body of the voice input unit 42.

This shape is a shape conforming to a conventional general receiver, and a shape that can be recognized as a conventional general receiver unless the receiver is excessively conscious of fashion.

The user of the microphone 40 grasps the handle 43 so that the earpiece 41 of the microphone 40 is placed on one of the left and right ears, and the voice input unit 42 is positioned near the mouth. , Fix. Next, the voice input content is input from the mouse piece 421 of the voice input unit 42 to a sound pressure / drive force conversion mechanism of a voice input device (not shown) that is a main body of the voice input unit 42, and a diaphragm and other components are input. A vibration is induced in the mechanical system consisting of: and the sound is converted into a sound signal by the electro-acoustic conversion mechanism. The sound signal is converted into a sound input target software such as an electronic medical record via the input code 44. Can be entered into the installed computer system. This speech signal is processed by speech recognition software installed in a computer, for example, through processes such as acoustic analysis → phone segmentation → phoneme recognition → syllable recognition → sentence recognition. Can be input to software. The speech recognition software can use a commercially available product or one that is uniquely programmed. Also, the software to which the speech recognition signal is to be input can be freely selected from existing software having a speech recognition function, and independently programmed software can be used.

As the voice input device of the voice input unit 42, for example, a carbon transmitter, an electrodynamic sound pressure microphone, an electrostatic sound pressure microphone, a unidirectional microphone, or the like is used. However, it is preferable to use a unidirectional microphone in order to improve the sensitivity of speech recognition as much as possible.

The earpiece 41 has a structure whose primary purpose is the function of making the microphone 40 a handset type. If only the shape requirements are satisfied, no other functions are added. In addition, it satisfies the requirements of the present invention. However, it is preferable to provide a speaker mechanism (sound output mechanism) capable of converting an audio signal into sound pressure inside the earpiece 41, like the earpiece of a conventional receiver. For example, the audio signal input and converted from the audio input unit 42 is sent to the computer system to which the audio is to be input, and at the same time, via the conductive cord provided inside the handle unit 43, By sending the sound pressure to the speaker mechanism and converting it to sound pressure, the voice input user can more clearly understand his / her input voice. Also, for example, an audio signal relating to an instruction of an audio input procedure or the like is output from a computer system to be audio input, and this is connected to a speaker mechanism of an earpiece through an output cord or the like, and the audio signal is output. Can be converted into sound pressure information and output, and the output contents can be transmitted to the voice input person, so that the handling of software such as electronic medical records can be simplified.

Note that the input code 44 and the output code used in the present embodiment can be made cordless by using a known cordless means, for example, wireless.

[0027]

[Effect of the invention]

 According to the present invention, a voice input means suitable for use of a voice recognition system or the like in an electronic medical record is provided.

[Brief description of the drawings]

FIG. 1 is a schematic diagram of a headset type microphone.

FIG. 2 is a schematic diagram of a fixed microphone.

FIG. 3 is a schematic diagram of a hand microphone.

FIG. 4 is a schematic view of an embodiment of the present invention.

Claims (3)

[Utility model registration claims] 1. An earpiece, a voice input part, and a handle part, wherein the earpiece and the voice input part are connected by a handle part, and the earpiece is placed on a human ear. A microphone for inputting a voice recognition system, wherein an earpiece and a voice input unit are provided in a handle unit such that the voice input unit is located near a human mouth. 2. A sound input from a sound input unit or a sound from a computer system,
The microphone for inputting a voice recognition system according to claim 1, further comprising a voice output mechanism capable of outputting a voice. 3. The microphone for inputting a voice recognition system according to claim 1, wherein the microphone for inputting a voice recognition system is a microphone for inputting a voice recognition system of an electronic medical record.