JPH09292896A - Vehicle navigation device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enhance the voice recognition rate in a vehicle navigation using voice input. SOLUTION: This device uses voice input for various input operations in navigation, stores the frequency of use of recognized words in a learning memory 6a in the past voice input, and specifies voice input words considering the frequency of use stored in the learning memory 6a to the degree of recognition of that word when voice input words are voicerecognized by means of a voice recognition device 8.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicular navigation device that guides a vehicle by using voice input from an occupant.

[0002]

In a vehicle navigation device, items to be selected are hierarchically displayed on the display screen in selecting a display map, setting a destination route, etc., and performing item selection. , So that desired settings can be made. In this case, various proposals have been made to select an item by voice input using voice recognition.

However, in this type of speech recognition, the recognition rate is a problem, and in a noisy environment such as a vehicle compartment, it may be difficult to recognize depending on the word to be recognized. The present invention has been made in view of the above problems, and an object thereof is to increase the voice recognition rate in vehicle navigation using voice input.

[0004]

In order to achieve the above object, in the invention according to claim 1, the frequency of use stored at the time of past speech input is set to the degree of recognition of a word recognized by the speech recognition means. In addition, the feature is that the word input by voice is specified. The items input by the occupant to provide driving guidance are used in a biased frequency depending on the occupant's age group and regional characteristics. Therefore, by adding the frequency of use to voice recognition, the voice recognition rate at the time of voice input can be increased.

According to the second aspect of the present invention, when there are a plurality of words whose recognition degree in consideration of the frequency of use is equal to or higher than a predetermined value, the occupant selects which of the words is voice input. Therefore, even if there are a plurality of similar words, it is possible to reliably identify the word that has been input by voice.

[0006]

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention shown in the drawings will be described. FIG. 1 shows the overall configuration of a vehicle navigation device showing an embodiment of the present invention. In this embodiment, a GPS receiver 1, a gyro 2 and a wheel speed sensor 3 are provided to detect the current position of the vehicle. G
The PS receiver 1 receives radio waves from an artificial satellite and outputs a signal indicating the current position of the vehicle. The gyro 2 outputs a signal according to the horizontal angular velocity of the vehicle. The wheel speed sensor 3 detects the traveling distance of the vehicle based on the rotation speed of the wheels and outputs a distance signal.

The map data storage device 4 includes a storage medium such as a CDROM for storing map data, and outputs the map data to the control device 6. The input device 5 has various input keys operated by an occupant of a vehicle and outputs various input signals necessary for navigation. The control device 6 is configured to include computer means such as a microcomputer, and executes various arithmetic operations. For example, in the navigation process, the GPS receiver 1, the gyro 2,
The current position is obtained from the signal from the wheel speed sensor 3, the map data is read from the map data storage device 4, and the current position and the road map are displayed on the display device 7. Further, when the destination is set and route guidance is performed, the route to the destination is displayed on the display device 7, and the voice generation device 9 generates a guidance voice according to the route. Regarding the operation input when performing navigation, in addition to the input from the input device 5, voice input by the voice recognition device 8 can be performed.

The display device 7 is provided in the instrument panel portion of the vehicle and displays a current location and a road map.
A liquid crystal display device, a CRT display device, or the like can be used as the display device 7. The voice recognition device 8 includes an operation switch that is operated when performing voice recognition. When the operation switch is operated, the voice generated by an occupant is taken in from a microphone installed near the driver's seat to perform voice recognition. Then, a signal indicating that the recognition has been performed is output to the control device 6, and a word having a high recognition level (all words having a recognition level equal to or higher than a predetermined value) and data indicating the recognition level are output to the control device 6. It should be noted that the voice recognition device 8 stores registration data of words for voice recognition, and performs voice recognition by comparing input voice data with the registration data.

Upon receiving the data for voice generation from the control device 6, the voice generating device 9 generates a voice corresponding to the data in the passenger compartment. The operation of the above configuration will be described. The control device 6 receives power supply from the vehicle-mounted battery and executes various arithmetic processes including the above-mentioned navigation process. In that case, those processes are executed using the operation by the input device 5 and the voice input.

There are various operations in that case. For example, a case of calling a map in the vicinity of a place name or a facility name will be described. First, the input device 5 is operated to display the map menu shown in FIG.
Here, when voice input of "map search" is performed for each item displayed on the menu, the menu shown in FIG. 2B is displayed. After this, the voices of "Station", "Tokyo", "Yamanote Line", and "Harajuku" are reproduced in Fig. 2 (c), (d), (e).
When you sequentially input in accordance with the menu display of, the map near Harajuku is finally displayed.

Such voice input can be used in various navigation operations other than the above. Next, a specific process of the above voice input will be described with reference to the flowchart shown in FIG. First, it is determined whether or not a signal indicating that recognition has been generated from the voice recognition device 8 (step 101). Since the voice recognition device 8 outputs a signal to that effect when voice recognition is performed, as described above,
The determination is made depending on whether or not the signal is generated.

When a signal indicating that the recognition has been made is generated from the voice recognition device 8, a word having a high recognition degree and data indicating the recognition degree are read from the voice recognition device 8 (step 1).
02). Next, the degree of recognition of these data is changed by the learning data stored in the learning memory (step 103). FIG. 4 shows the data structure of the learning memory.

The class memory is a memory (a power-backed-up one or a semiconductor-type non-volatile memory) 6a which is stored in the controller 6 in a non-volatile manner, and stores the words recognized in the past and the number of times of recognition. It consists of the counter shown. In step 103, when the recognized word is stored in the learning memory, the recognition degree of the recognized word is
The taunt value of the counter for that word is multiplied by a certain constant and added. That is, the recognition frequency of the recognized word is changed by adding the past usage frequency.

Thereafter, a word having a degree of recognition equal to or higher than a predetermined value is selected (step 105), and it is determined whether or not there are two or more selected words (step 105). When there are two or more selected words, the display device 7 displays the selected words in a list, and the occupant selects which of the words is input by voice (step 106).

The selected word and the word in which the number of words whose recognition degree is equal to or higher than a predetermined value are one are specified as the voice input word and used for the voice input operation. Further, the counter stored in the learning memory is counted up for the specified word (step 107) and reflected in the voice input process from the next time.

Therefore, according to the above-described embodiment, in the voice input for the navigation operation, the voice input word is obtained by adding the recognition frequency of the recognized word to the use frequency stored in the past voice input. Therefore, the voice input of the occupant can be learned and the voice recognition rate can be increased. That is, in navigation operations, the items operated by the occupants are used in an uneven frequency depending on the age group, regional characteristics, etc. of the occupants, and the voice recognition rate is increased by adding the frequency of use to the voice recognition. You can

[Brief description of drawings]

FIG. 1 is an overall configuration diagram of a vehicular navigation device showing an embodiment of the present invention.

FIG. 2 is a flowchart showing a voice input process of a control device 6 in FIG.

FIG. 3 is a diagram showing a data configuration of a learning memory.

FIG. 4 is an explanatory diagram for explaining a voice input operation.

[Explanation of symbols]

1 ... GPS receiver, 2 ... gyro, 3 ... wheel speed sensor,
4 ... Map data storage device, 5 ... Input device, 6 ... Control device, 6a ... Learning memory, 7 ... Display device, 8 ... Voice recognition device, 9 ... Voice generating device.

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI technical display location G08G 1/0969 G08G 1/0969

Claims (2)

[Claims] 1. A vehicle navigation device that guides a vehicle using voice input from an occupant, wherein a voice recognition unit (8) for voice-recognizing the voice-input word, Use frequency storage means (6a) for storing the use frequency of the word specified as the voice input, and the recognition degree of the word recognized by the voice recognition means,
Means (103 to 106) for identifying the voice-inputted word in consideration of the stored usage frequency, and means (107) for updating the stored usage frequency for the identified word. And a vehicle navigation device. 2. The means for identifying a word input by voice, when there are a plurality of words having a degree of recognition in consideration of the frequency of use or more, which word has been input by voice by an occupant. Means for selecting (105, 106)
The vehicle navigation device according to claim 1, further comprising: