JPH0337197B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention provides a system for commanding the operation of a load in response to a voice command by recognizing voice commands based on registered voice data of an operator registered in advance. The present invention relates to a voice recognition device for a vehicle.

Conventionally, as this type of speech recognition device, there is one shown in FIG. 1, for example.

In FIG. 1, 1 is a microphone installed near the driver's seat. When using the voice recognition device, first register a predetermined voice command by the operator's voice by operating a registration switch 2. . That is, when the registration switch 2 is turned on, the switch input detection circuit 4 detects the switch operation and produces an output on the signal line 4a, and the control section 5 gives a registration mode processing command to each circuit section. In such a registration mode, a predetermined voice command, for example, a voice command "radio" specified for turning on a car radio, is transmitted to the microphone 1.
When the voice command is issued to the user, the microphone 1 converts the voice command into an electrical signal, which is amplified by the voice input circuit 6. This voice command signal is input to the voice detection circuit 7, and the voice command signal is started as a time-series signal. The time point is detected based on a rising change in the signal level, and the controller 5 is instructed to start the registration process.

Therefore, the voice command signal from the microphone 1 is divided into predetermined frequencies by the band filter group 8.
In the parameter extracting section 9, the signal is squared or rectified to obtain the audio power spectrum in each frequency band, digitally converted into time series data serving as audio power information, and stored in the memory 10. At this time, since the registration mode is set by operating the registration switch 2, the voice data in the memory 10 is transferred to the registration data storage section 11 and registered as reference data for performing voice recognition processing.

Next, after registering the voice command, if you want to listen to the onboard load, for example, the radio, turn on the command start switch 3 to give the predetermined voice command "radio" to the microphone 1. .

In response to this ON operation of the command start switch 3, the switch input detection circuit 4 produces an output on the signal line 4b, and the control section 5 gives a recognition mode control command to each circuit section. The voice commands input from 1 are converted into voice data and written into the memory 10 in the same way as at the time of registration. The audio data stored in this memory 10 is
A similarity comparison processing section 12 calculates the degree of similarity between a plurality of registered data sequentially read out from the registered data storage section 11. To calculate this degree of similarity, after normalizing the time axis and normalizing the level of the audio data and the registered data, the degree of similarity is determined by performing calculations on Tievishiev distance and other distances. Become.

The similarity value calculated by the similarity comparison processing section 12 is judged by the recognition word processing judgment section 13 as to whether or not the similarity value is within a predetermined threshold range.
When it is within the range, a command corresponding to the audio data having this degree of similarity is output to the drive unit 14, and the car radio is turned on.

However, in conventional devices, the gain of the amplifier is constant, so the gain must be set to a small value in order to enable recognition even under high noise conditions. For this reason, voice registration, which is often done in a relatively quiet place, requires a fairly loud voice.For this reason, if the operator uses force to speak, the registered pattern may differ greatly from the pattern used for recognition. There is a possibility that it may not be recognized.

The present invention has been made by focusing on such conventional problems, and it registers a predetermined voice command inputted into a microphone when a registration switch is turned on as recognition reference data, and then registers a predetermined voice command inputted into the microphone after this registration. In a device that commands the operation of a load corresponding to a voice command based on the above registered data, the voice command input through a microphone is processed with a larger amplification gain during voice registration than during voice recognition in order to improve the recognition rate. It is designed to amplify it.

The present invention will be explained below based on the drawings.

FIG. 2 is a block diagram showing one embodiment of the present invention.

First, to explain the configuration, 1 is a microphone that converts voice commands from the operator into electrical signals, 2 is a registration switch that is turned on when registering voice commands,
3 is a command start switch that is turned on when the load is activated by a voice command after registration, and 4 is a switch signal that is turned on when the registration switch 2 or command start switch 3 is turned on, and the registration switch 2 is connected to the signal line 4a. Regarding the command start switch 3, a switch input circuit 5 outputs an output to the signal line 4b, and a switch input circuit 5 outputs an H level output to the signal line 5a at the time of registration based on the signal from the switch input circuit 4, and outputs an H level to the signal line 5b at the time of command start. This is a control section that generates level output. Reference numeral 6 denotes an audio input circuit that amplifies the electric signal given from the microphone 1 and outputs it as a voice command signal.This audio input circuit 6 includes an amplifier 61 having different amplification gains _G1 and _G2 ,
62 and an analog switch 63 that switches the outputs of the amplifiers 61 and 62 based on the output of the control circuit 5. The amplification gain _G2 is 1.2 to 1.8 times that of _G1 , preferably 1.6 times. Amplifier 61,6
The amplification gain of 2 is set in a relationship such that G ₁ <G ₂ , and when the control section 5 generates an H level output on the signal line 5b by turning on the command start switch 3,
The analog switch 63 is switched as shown in the figure to take out the output of the amplifier 61, and when the registration switch 2 is turned on, the output to the signal line 5a of the control section 5 causes the analog switch 63 to switch as shown by the broken line. The output of amplifier 62 is taken out. That is, at the time of registration, the amplifier 62 having an amplification gain of G ₂ is selected, and when a voice command is issued after registration, the amplifier 61 with the amplification gain of G ₁ is selected, and therefore the amplifier 62 having an amplification gain of G 1 is selected. At the time of a voice command, the amplification gain of the electrical signal from the microphone 1 is made larger during registration than during recognition. Note that the circuit section that registers voice commands based on the output of the voice input circuit 6 and recognizes the voice commands after registration has the same configuration as the conventional device shown in FIG. 1, so a description thereof will be omitted. There is.

Next, the operation of the embodiment shown in FIG. 2 will be explained.

When registering a voice command, first turn on the registration switch 2. By turning on the registration switch 2, the switch input circuit 4 generates an output on the signal line 4a, and therefore the control unit 5 sets the signal line 5a to H level and switches the analog switch 63 in the audio input circuit 6 as shown by the broken line. , amplifier 62
The output will be extracted. Therefore, if the operator issues a predetermined voice command to the microphone 1 for registration, the microphone 1 converts it into an electrical signal, which is amplified by the amplifier 62 with an amplification gain of G ₂ , and then sent to the analog switch. 6
3 as a voice command signal for registration, and similarly to the conventional device, the signal is converted into voice power data after frequency division and registered as reference data for voice recognition in the registration data storage unit.

After registering such voice commands, in order to operate the load by voice commands, if the command start switch 3 is operated to issue a predetermined voice command, the control section based on the output of the signal line 4b of the switch input circuit 4 5 (signal line 5b), the analog switch 63 is switched as shown in the figure, and the amplifier 61 converts it into an electrical signal by the microphone 1, and the amplification gain is increased.
_G1 is amplified and output, and through recognition processing of this voice command signal and pre-registered registration data, a command is issued to operate the load corresponding to this voice command.

FIG. 3 shows another embodiment of the audio input circuit 6 in the embodiment _of FIG _. The amplification gain of the electrical signal from the microphone 1 is changed by switching the analog switch 6. Similarly to the embodiment shown in FIG.
3 is switched as shown by the broken line to amplify the electrical signal from the microphone 1 with the amplification gain determined by the feedback resistor R ₃ and the input resistor R ₁ , and when the load is activated by a voice command, the analog switch 63 is switched as shown by the solid line, and the feedback resistor _R2 and the input resistor
The electrical signal from microphone _{1 is amplified with the amplification gain determined by R 1} , and the amplification gain at the time of registration is set to be larger than the amplification gain at the time of voice command, so even at the time of registration, the voice cannot be heard with normal utterance. Can be registered.

Next, voice registration when the voice recognition device according to the present invention is realized using a microcomputer will be explained with reference to FIGS. 4 and 5.

Figure 4 shows a speech recognition device using a microcomputer, and is composed of a microphone 1, an amplifier 20 that amplifies the voice command input to the microphone 1 with a predetermined amplification gain, and a group of bandpass filters 21. The voice processing section 22 is an A/D converter for inputting voice commands to the microcomputer 24, and 23 is an operation switch for registering voice and switching recognition.

The speech recognition device shown in FIG. 4 will be explained with reference to the flowchart shown in FIG.

When the voice registration or voice recognition operation switch is pressed, the microphone enters an input waiting state, and it is further determined in step a whether or not the registration switch has been pressed. If the registration switch is not pressed, the amplifier is set to a predetermined amplification gain in step c. If the registration switch is pressed, step b sets the amplification gain larger than the predetermined amplification gain, and the voice command input from the microphone 1 is amplified by the set amplification gain and then processed by a bandpass filter or the like. The audio data is then A/D converted in step d, and the audio data is sequentially stored in step e. This continues until the end of the voice according to step f. After the voice input is completed, it is determined in step g whether or not the registration switch is pressed again.
If it has been pressed, registration processing such as setting the audio data as a registration reference pattern has been performed in step h. If the recognition switch has been pressed, recognition processing such as comparing the voice data with a registered reference pattern stored in advance is performed in step i.

As explained above, according to the present invention, the voice command input through the microphone is amplified with a larger amplification gain during voice registration than during voice recognition, so the operator who operates the voice command during voice registration is the same as usual. The recognition rate can be improved because the registered pattern will not change if the command is uttered with force.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an example of a conventional speech recognition device, and FIG. 2 is a block diagram showing an embodiment of a speech recognition device according to the present invention in which the amplification gain of the speech signal is lowered at the time of registration. 3 is a circuit diagram showing another embodiment of the variable gain amplifier used in FIG. 2, FIG. 4 is a block diagram when the speech recognition device according to the present invention is realized by a microcomputer, and FIG. This is a flowchart used in 1...Microphone, 2...Registration switch, 3
...Command start switch, 4...Switch input detection circuit, 5...Control unit, 6...Audio input circuit, 7...
...Audio detection circuit, 8...Band filter group, 9...
Parameter extraction circuit, 10... Memory, 11...
Registration data storage unit, 12...similarity comparison processing unit,
13... Recognized word judgment processing unit, 14... Drive unit, 2
0, 61, 62, 60...Amplifier, 21...Audio processing unit, 22...A/D converter.

Claims (1)

[Claims] 1 Register a predetermined voice command input when turning on the registration switch as recognition reference data, recognize the voice command input after this registration based on the registered data, and operate the load corresponding to the voice command. Vehicle use, characterized in that the commanding device is provided with registration data processing means for amplifying the voice command input into the microphone with a larger amplification gain during voice registration than during voice recognition and converting it into the registered data. Speech recognition device.