JPH033539A - Voice command type automobile telephone set - Google Patents

Abstract

PURPOSE:To improve the convenience of use by collating a command voice with a registered voice, searching a possible voice among registered voice signals when dissident, displaying and selecting the voice. CONSTITUTION:A voice recognition computer 3 receives a command voice signal inputted from a microphone 1 when a registration switch 5 is set, registers the relevant voice information into a memory and receives the voice signal from the microphone 1 when a command switch 6 is closed, The computer 3 collates the received command voice information with the registered voice information and outputs and displays a prescribed registered content onto a display section 4 when they are coincident. When dissident, all possible voice signals (similar voice) are sought among the registered voice signals and the result is displayed on the display section 4. Thus, the command voice with high recognition rate is selected more than the normal command voice among them. Thus, convenience of use is improved.

Description

[Detailed description of the invention] (Industrial application field) The present invention relates to a voice command type car telephone.

(Prior Art and Problems to be Solved) When a driver communicates with the other party using the handset of a car phone while driving a car, it may be a problem for safe driving if the driver is driving the car and using the phone at the same time. However, it would be desirable for the driver to be able to operate the telephone without having to pick up the handset.

Therefore, we have a regular handset phone and a hands-free phone with a speaker and microphone placed on the instrument panel in front of the driver's seat, and when the driver uses the phone, he can switch to the hands-free phone.
2. Description of the Related Art Car telephones are known in which the driver selects and calls the other party's station by voice, and communicates with the other party using a microphone and a speaker.

A so-called voice-command car phone that uses voice to select or call the other party searches for the one that matches or most closely resembles the command voice from among a large number of registered voices, and displays the corresponding voice if it is found. If there is no command, for example, the message "Please try again" is displayed and the driver is requested to speak the command again.

However, at present, the recognition rate of command voices is at most 80 to 90%, and in some cases the command voice may be requested many times, making it difficult to use.

The present invention has been made in view of the above-mentioned points, and provides a voice-command car phone in which commands are given in stages, and the command is first accepted using a general voice, and then accepted using a method with a higher recognition rate. The purpose is to provide.

(Means for Solving the Problems) In order to achieve the above object, according to the present invention, a command voice is inputted, and the registered voice is checked against a registered voice, and when a match is found, the registered contents are displayed. In a command-type car phone, the command voice is checked against the registered voices, and if they do not match, all possible voices are searched and displayed from among the registered voices, and the user is prompted to select from among the displayed voices. The structure is equipped with means.

(Operation) At first, information is input using a general command voice, and when the command voice is checked against the registered voice and it does not match, all possible ones are searched from among the registered voices. Display. Then, the user is prompted to select a voice or the like with a higher recognition rate from among the displayed ones.

(Example) An embodiment of the present invention will be described in detail below with reference to the accompanying drawings.

FIG. 1 shows a voice input section of a voice command type car phone to which the present invention is applied, and a microphone 1 is placed, for example, at a predetermined position in front of the driver on an instrument panel (not shown). It is said that it is possible to input voice well. The output signal of this microphone l is amplified by an amplifier 2 and then input to a voice recognition computer 3. The output signal of this voice recognition computer 3 is input to the display section 4. One end of each of the registration switch 5 and the command switch 6 is grounded, and the other end thereof is connected to the input side of the voice recognition combiator 3, respectively.

When the registration switch 5 is on, the voice recognition computer 3 receives a command voice signal input from the microphone 1 and registers the voice information in the memory, and when the command switch 6 is on, it does not accept the voice signal from the microphone 1. wear. The computer 3 compares the received command voice information with the registered voice information, and if it loses, outputs the registered predetermined content and displays it on the display unit 4, and if they do not match, the computer 3 checks the registered voice information. Search out all the possible voices (those whose voices are similar) from the currently available voices and display them on the display section 4, and select from among them, for example, "Ichi", which has a higher recognition rate than the ordinary command voice. , ``2'', ``san'', etc. are selected by voice commands. In addition, when selecting by voice, select tR.
As mentioned above, the voice to be used is determined to have a high recognition rate.

The operation will be explained below with reference to the flowchart shown in FIG.

When registering the other party's name in the voice recognition computer 3, the registration switch 5 is turned on (step l) and the other party's name is uttered into the microphone 1. The computer 3 receives the audio information output from the microphone 1 and amplified by the amplifier 2 (step 2), and registers it (step 3). In this way, the names of the required parties are registered in the computer 3 in sequence. do.

If the driver wants to make a phone call while driving, he turns on the command switch and then speaks the other party's name, for example, rsAITOJ, into the microphone l. The computer 3 determines whether or not the command switch 6 is turned on (step 4), and if the determination answer is negative (No), the process returns to step 1, and if the determination answer is affirmative (YES), the command voice input from the microphone 1 is output. is accepted (step 5), and the computer 3 compares the accepted command voice with the registered voice (step 6) to determine whether they match (step 7).

Computer 3 determines that the discrimination answer in step 7 is affirmative (YES).
), the registered contents, for example, the other party (SA
ITO) telephone number, etc. is displayed on the display unit 4 for the driver to confirm, and if the answer is negative (No), possible voices are selected from among the registered voices (those whose voices are similar). are all searched and displayed on the display section 4 (step 9). For example, 1...5ATO12...5AITO13...
- Display as ITO etc. (step 10). Note that this information may be made to be recognized by the driver by voice.

Next, the driver selects a desired name from among the displayed names using a voice with a high recognition rate. For example, say ``Two!

, and displays the registered contents on the display section 4 (step 8).

In this way, the Komoid voice is divided into stages, and the voice that is easy to recognize is accepted at first, and if it is checked against the registered voice and there is no match, the voice with a higher recognition rate is accepted at the next stage. . It is also possible to increase the number of selection stages (gates) depending on the number of registered voices; for example,
If 20 to 50 people are registered, use 2 gates, and if 50 to 100 people are registered, use 3 gates to ensure good recognition without deteriorating usability. be able to. Incidentally, the second and subsequent selections may be made by other means, such as a switch, instead of by voice.

(Effects of the Invention) As explained above, according to the present invention, in a voice command type car phone, in which a command voice is inputted, the voice command is compared with a registered voice, and when a match is found, the registered contents are displayed. The command voice is compared with the registered voice, and if they do not match, the command voice is searched for and displayed from among the registered voices, and the voice is selected from among the displayed voices. By doing this,
This eliminates the trouble of repeatedly uttering a command voice with a low recognition rate over and over again, and has the effect of making it easier to use a voice command type telephone.

4,

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of the voice input section of a voice command type car telephone according to the present invention, and FIG.
3 is a flowchart showing the operation of the voice input section shown in the figure. 1...Microphone, 2...Amplifier, 3...Voice recognition computer, 4...Display section, 5...Registration switch, 6...Command switch.

Claims (1)

[Claims] In a voice command type car phone that inputs a command voice and checks it against the registered voice and displays the registered contents when it matches, if the command voice and the registered voice do not match. A voice command type car telephone is characterized in that it is equipped with a means for searching and displaying all possible voices from registered voices and for making a selection from among the displayed voices.