JPS58190996A - Voice recognition system for vehicle - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a voice recognition system for a vehicle, and a dual-purpose voice recognition system that recognizes voice commands of a person speaking and operates in-vehicle equipment instructed by the voice commands.

In recent years, vehicles such as automobiles are equipped with a large number of various stand-alone devices such as air conditioners and audio equipment. Therefore, it is necessary to easily and reliably operate these in-vehicle devices without disturbing the driver's concentration, and various operating methods have been developed and put into practical use.

One of the operating methods for such in-vehicle devices is a system that uses a voice recognition device to operate the in-vehicle devices, in which the driver can operate the in-vehicle devices by simply speaking voice commands into the microphone. can be done. Therefore, even when the driver operates on-vehicle equipment while driving, the driver can do so without taking his/her line of sight away from the direction in which the vehicle is traveling, which is extremely effective for safe driving.

However, voice recognition devices are greatly affected by noise, and can misinterpret voices from other people sitting in the front passenger seat or back seat, or voices coming from the radio, as voice commands, leading to erroneous operation of in-vehicle devices. There are many things that happen.

Such erroneous operation of in-vehicle equipment is not only troublesome for the driver, but also poses problems for safe driving depending on the type of erroneously operated in-vehicle equipment, so effective countermeasures are required.

The present invention has been made in view of such conventional problems, and its purpose is to provide a voice recognition method for vehicles that effectively prevents erroneous operation of in-vehicle equipment due to speech by speakers other than the registered specific speaker. I have a lot to offer.

To achieve this objective, the method of the present invention is a vehicle voice recognition method that recognizes a speaker's voice commands and operates in-vehicle equipment instructed by the voice commands. 6. Register the speaker's voice command, temporarily store it in memory, and compare the voice pattern vector of the stored voice command with the registered pattern vector to determine whether the speaker is a registered specific speaker or not. It is characterized in that the voice commands stored in the memory are continued and the voice recognition is performed only when the voice command is recognized as a specific speaker.

Next, preferred embodiments of the present invention will be described based on the drawings.

FIG. 1 shows a flowchart showing an embodiment of the speech recognition method of the present invention.

A characteristic feature of the present invention is that in a vehicle voice recognition system that recognizes a speaker's voice commands and operates in-vehicle equipment instructed by the voice commands, prior to voice recognition of the speaker's voice commands, The purpose of this system is to perform speaker recognition to determine whether or not the speaker is a registered specific speaker, thereby preventing erroneous operation of in-vehicle equipment due to utterances from a person other than the registered speaker.

Therefore, in the present invention, the voice pattern vector of a specific speaker is registered in advance.

Then, the voice command uttered by the speaker is temporarily stored in a memory, and the voice pattern vector of the stored voice command is compared with the registered pattern vector to perform speaker recognition to determine whether or not the speaker is a registered specific speaker. When the speaker is recognized as a specific speaker, the voice commands stored in the memory are read out again and voice opening is performed.

As described above, according to the present invention, only the voice commands of registered speakers are selected through speaker recognition, and only the voice commands of these voice commands are recognized. Misoperation can be reliably prevented.

FIG. 2 shows a vehicle voice recognition device that performs voice recognition according to the flowchart shown in FIG.

In the embodiment, the speaker's voice is first input through the microphone 10 . Then, this audio signal is transmitted to the amplifier 12,
The A/D converter 18 is input via the filter puncture 14 and the analog multiplexer 16.
The signal is converted into a digital signal by the converter 18 and input to the CPU 20 .

The CPU 20tl;t thus digitized and manually inputted audio signals are stored in the ROM 22 and subjected to arithmetic processing according to the mourning program. Since the ROM 22 stores a program according to the flowchart shown in FIG. 1, the input voice signal is processed according to the voice recognition method of the present invention.

In the device of the embodiment, in order to perform the speaker recognition, a voice pattern vector of a specific speaker is registered in the RAM 24 in advance.

This registration operation is performed by the CPU 2G by a control device (not shown).
The registered mode pressure is controlled, and the utterance of a specific speaker (usually a predetermined single @ or sentence) is transmitted to microphone 1.
This is done by inputting 0. The voice signal of the specific speaker input in this way is converted into a digital signal and input to the CPU 20 which is controlled to be in the registration mode. C
P'U20 first acoustically analyzes the input audio signal,
After converting the signal into a series of parameters that more clearly represent the characteristics of the speaker, this signal is further subjected to time-axis averaging, segmentation/
A speech pattern vector of a specific speaker is formed by performing pattern conversion based on the speech pattern. This voice pattern vector is then registered in the RAM 24.

In this manner, when the voice pattern vector of a specific speaker is registered in the RAM 24, the indicator light is lit to indicate that the next voice recognition is possible.

Next, a control device (not shown) controls the second CPU 20 to a voice recognition mode, and in this state, when the speaker inputs a voice command to the microphone 10 to instruct the operation of a predetermined in-vehicle device, this voice signal is transmitted to the A/D. It is converted into a digital signal by converter t8 and temporarily stored in RAM24.

Further, the ready display is stopped at the same time as this voice command is input.

The CPU 20 sequentially reads out the voice commands temporarily stored in the RAM 24 and performs predetermined acoustic analysis and pattern conversion on the voice commands to form a voice pattern vector. Then, the speaker's voice pattern vector formed in this way is compared with the voice pattern vector of the specific speaker registered in advance in the RAM 24 to determine whether the speaker is registered and is a good specific speaker. Speaker recognition is performed.

As a result of such speaker recognition, if it is determined that the speaker is not a registered specific speaker, the CPU 20 lights up a reject lamp for 7 seconds to display a message that the vehicle equipment will not be operated.

Further, if the speaker is recognized as a registered specific speaker as a result of the speaker recognition, the CPU 20 stores the information in the RAM 24.
The voice command temporarily stored in the KRAM 24 is read out again, and this voice command is compared with registered words already stored in the KRAM 24. In the embodiment, since a plurality of registered words corresponding to the operations of various in-vehicle devices are stored in the RAM 24, the CPU 20 determines whether the voice command read from the FiRAM 24 matches any one of the plurality of registered words. Perform voice recognition.

As a result of this voice recognition, if the voice command is recognized to match any registered word, the CPU 20 outputs a control signal via the interface 26 to operate the in-vehicle device 28 corresponding to the instruction content of the voice command. Do this.

Also, audio la! As a result of the recognition, if the voice command does not match the pre-registered voice pattern, the CPU 20ii
The reset lamp will be lit for 7 seconds to indicate that the on-vehicle equipment will not be operated.

Since the sound recognition device of this embodiment has the above configuration, for example, the driver's voice pattern vector is registered in advance, and the microphone 10 is installed at a position where the voice of the driver seated in the driver's seat is input. By doing so, only the voice commands uttered by the driver can be reliably recognized and the desired in-vehicle equipment can be operated.

In addition, even if the microphone 10 is mixed with the voices of speakers seated in the front passenger seat or rear seats, or other noises, the mixed audio signals are eliminated by speaker recognition. erroneous operation is reliably prevented.

Furthermore, if voice pattern vectors are registered in advance for a plurality of people driving the vehicle, it is possible to reliably operate the in-vehicle equipment of the vehicle even if the driver changes.

In addition, if the voice recognition method of the present invention is used to unlock the locking mechanisms installed in the console box, trunk, etc. of the pet, only those who are registered speakers will be able to unlock these locking mechanisms. . Therefore, there is no need for a key to unlock the locking mechanism, and moreover, the locking mechanism can be easily unlocked by simply uttering a voice command.

As explained above, according to Hokomeitara, only the voice uttered by a registered specific speaker is recognized, and other voices are not recognized. Misoperations are reliably prevented. Therefore, the present invention enables accurate operation of in-vehicle equipment without sacrificing the advantages of the voice recognition method.

[Brief explanation of the drawing]

FIG. 1 is a flowchart showing a preferred embodiment of the present invention, and FIG. 2 is a block diagram of a vehicle speech recognition device that performs speech recognition. In-vehicle equipment.

Claims (1)

[Claims] +11 In a vehicle voice recognition system that recognizes the speaker's voice commands and operates the in-vehicle equipment instructed by the voice commands, a voice pattern vector of a specific speaker is registered in advance, and the voice commands of the speaker are registered in advance. is temporarily stored in the memory, and the voice pattern vector of the stored voice command is compared with the registered pattern vector to perform speaker recognition to determine whether or not the speaker is a registered specific speaker, and is recognized as a specific speaker. A voice recognition system for a vehicle, characterized in that a voice command stored in a memory is read out and voice recognition is performed only when the voice command is received.