JPH06202690A - Acoustic equipment controller by speech recognition - Google Patents

Abstract

PURPOSE:To prevent malfunction due to speech misrecognition. CONSTITUTION:A speech recognition unit 2 once recognizing that the speech signal inputted from a microphone 1 is a predetermined command word outputs a signal corresponding to the command word to an interface 3. The interface 3 receives the signal from the speech recognition unit 2 and outputs an operation command to acoustic equipment 4, but when it is judged that the recognized command word is inadequate for the current operation state of the acoustic equipment 4, the current operation state of the acoustic equipment is maintained and the speech recognition unit 2 is restored to the state before the current recognition of the speech.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for controlling an audio device by voice recognition, and more particularly to a means for preventing malfunction due to voice recognition error.

[0002]

2. Description of the Related Art Conventionally, it has been proposed to operate a vehicle-mounted audio device by voice because it is difficult to operate it by hand while driving. In this case, the generated voice is picked up by the microphone, and the voice recognition unit determines the similarity of the frequency distribution etc. with the predetermined command word to determine which command word the generated voice is. To do.

When the voice recognition unit recognizes a voice as a specific command word, a signal corresponding to the command word is sent to the audio equipment, and the audio equipment operates according to the signal.

However, human voices do not always have the same pronunciation, and different people have different characteristics. In addition, the microphone collects environmental sound together with the voice, so that it is affected by the frequency distribution. Therefore, the voice recognition unit may erroneously recognize.

In order to prevent such erroneous recognition, in the voice load drive device proposed in Japanese Patent Publication No. 63-29755, a voice switch is pressed to reduce the volume level of the audio equipment when inputting a voice command word. Also, a means for closing the window to reduce the noise outside the vehicle is used.

Further, in the voice recognition device of the utility model registration application filed by the applicant of the present application (Japanese Patent Application No. 3-93143), a keyword and an individual word pattern for each speaker are stored in a keyword memory section. It is stored in the individual memory section. When the speaker first inputs a keyword, the pattern is compared with the pattern in the keyword memory section to determine the speaker and open the individual memory section of the speaker,
The input voice is recognized by the pattern stored in the opened individual memory unit.

[0007]

As described above, various measures have heretofore been proposed for preventing erroneous voice recognition in a voice recognition device for controlling a vehicle-mounted audio device or the like, but they are still complete. However, it has been desired to further improve the recognition accuracy.

The present invention has been made in view of the above points, and an object thereof is to provide a means for preventing erroneous voice recognition by a method different from the conventional one.

[0009]

SUMMARY OF THE INVENTION An audio equipment control apparatus for voice recognition according to the present invention outputs a signal corresponding to a command word when it recognizes that an input voice is a predetermined command word. And an audio device control device including an interface for receiving a signal from the voice recognition unit and outputting an operation command to the audio device, the recognized command word is inappropriate for the current operating state of the audio device. When it is determined that the current operation state of the audio device is maintained, the state before the recognition of the command word is restored.

[0010]

The audio device has various operating states such as an FM receiving state, an AM receiving state, a CD playing state, a tape cassette playing state, and an operation peculiar to each state. For example, operations such as fast-forwarding and rewinding the tape are operations that are performed only when the tape deck is in use, and a command word for performing such operations when the user recognizes that the user is in the FM reception state. Does not pronounce.

According to the audio equipment control apparatus of the present invention, the signal of the command word which is inappropriate for each operation state is stored in the interface in advance, and it is judged that such a signal is received from the voice recognition unit. The interface keeps the audio device in the current operating state as it is and restores the state before the voice is input to the voice recognition unit, thereby preventing malfunction due to voice misrecognition.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An audio equipment control apparatus using voice recognition according to an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing an audio equipment control apparatus according to an embodiment of the present invention.

Reference numeral 1 shown in the figure is a microphone, which converts a voice of a user of an acoustic device into an electric signal and recognizes the voice.
Output to. The voice recognition unit 2 is an interface 3
Controlled by, the pattern of the signal input from the microphone 1 is compared with the pattern of the command word registered in advance, which command word is judged from the similarity, and the signal corresponding to the command word is determined. Output to interface 3.

The interface 3 has a microcomputer 3a and a storage device, controls the voice recognition unit 2, analyzes and judges the signal input from the voice recognition unit 2, and outputs an operation command to the audio equipment 4. .

The audio device 4 is a CD player, an FM receiver,
It has a tape deck or the like and is operated by the voice recognition unit 2 or the key input device 6 to output a voice signal to the output device 5. The output device 5 is composed of a speaker or the like, and converts the input voice signal into sound.

Next, the operation of the microcomputer 3a will be described with reference to FIG. When the interface 3 becomes operable by turning on the power, the present operating state of the audio equipment is confirmed in step S1.

Next, in step S2, it is judged whether or not a signal for recognizing the voice is output from the voice recognition unit 2. If the signal is output, the process proceeds to step S3, and if not, step S1. Is repeated.

In step S3, the input signal is analyzed and the process proceeds to step S4. In step S4, it is determined whether or not the operation command of the input signal is appropriate for the current operating state of the audio device.
If not appropriate, the process proceeds to step S5.

In step S6, an operation command corresponding to the input signal is output to the audio equipment 4 to cause the audio equipment 4 to execute a state transition operation, and the operation proceeds to step S1.

In step S5, the voice recognition unit 2 is returned to the state before recognizing the voice this time while keeping the current state of the audio equipment 4, and the process proceeds to step S1.

[0021]

According to the audio equipment control apparatus by the voice recognition of the present invention, when the sound in which the operation is inappropriate is recognized from the current state of the audio equipment, the state before the voice recognition is restored. It is possible to prevent malfunction due to erroneous voice recognition.

[Brief description of drawings]

FIG. 1 is a block diagram showing an audio device control apparatus according to an embodiment of the present invention.

FIG. 2 is a flowchart showing an operation of the audio device control apparatus.

[Explanation of symbols]

 1 Microphone 2 Voice Recognition Unit 3 Interface 3a Microcomputer 4 Acoustic Equipment 5 Output Device 6 Key Input Device

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁵ Identification code Internal reference number FI Technical display location G11B 15/10 C 9198-5D 19/16 B 7525-5D

Claims (1)

[Claims] 1. A voice recognition unit that outputs a signal corresponding to the command word when it is recognized that the input voice is a predetermined command word, and an audio device is operated by receiving a signal from the voice recognition unit. In an audio device control device including an interface that outputs a command, when it is determined that the recognized command word is inappropriate for the current operating condition of the audio device, the current operating condition of the audio device is maintained and the command word is maintained. A device for controlling audio equipment by voice recognition configured to return to the state before recognition.