JPH02202253A - Telephone set - Google Patents

Abstract

PURPOSE:To prevent dialing to other parties than the destination and to improve the reliability by comparing a numeral string outputted from a voice recognizing means with a telephone number stored in a storage means and outputting the name of destination corresponding to the telephone number in a synthesized voice when the coincident telephone number is obtained. CONSTITUTION:Numeral voice of a telephone number uttered by the operator is converted into a numeral with a high similarity and stored in a storage section 5. A comparison section 7 collates and compares the numeral string of the telephone number stored in the storage section 5 as the result of recognition with a subscriber number stored in advance in a storage section 8 and outputs a registration number corresponding to the subscriber number to a destination name output section 7b. A voice synthesis section 9 converts the number into a synthesized voice by a prescribed rule voice synthesis method and informed to the operator as a voice message expressing its character string through a speaker 10. Thus, an error of the inputted telephone number itself and the presence of an error of voice recognition are confirmed, resulting that the dialing to other parties than the destination is prevented.

Description

[Detailed description of the invention] [Purpose of the invention] (Industrial application field) The present invention relates to a telephone equipped with a voice recognition function.

(Prior Art) In recent years, telephones have been known that are equipped with a voice recognition function that recognizes a numeric string of a telephone number uttered by an operator, and that automatically make a call to a desired destination based on the telephone number that is the recognition result.

By the way, in a phone equipped with such a voice recognition function, after obtaining the digit string of the phone number as a recognition result, it immediately starts making a call using that phone number, so the phone number itself that is input by voice cannot be used. There is an error in
If an error occurs in the voice recognition process, a call may be made to an unintended destination, which significantly reduces the reliability of the telephone.

(Problems to be Solved by the Invention) As described above, in conventional telephones equipped with a voice recognition function, errors in the telephone number itself entered by voice (due to the operator's misremembering of the telephone number, etc.), There has been a problem in that an error in voice recognition processing may result in a call being made to an unintended destination.

The present invention is intended to solve such problems, and it is possible to check whether there is an error in the input telephone number itself or an error in voice recognition from the destination name output in synthesized voice. The purpose of this is to provide a telephone that can improve reliability by preventing calls from being made to unintended destinations.

[Structure of the Invention] (Means for Solving the Problems) In order to achieve the above-mentioned object, the telephone of the present invention includes a storage means in which a plurality of destination names and their corresponding telephone numbers are stored in advance, and a telephone. a first voice recognition means that recognizes a voice associated with the number and outputs a string of numbers of the telephone number; and a comparison means that compares the string of numbers outputted from the voice recognition means with the telephone number stored in the storage means. and a first synthesized speech output means for outputting a destination name corresponding to the telephone number in synthesized speech when a telephone number matching the number string is obtained as a result of the comparison by the comparison means; result,
When a phone number matching the above number string is not obtained,
a second synthesized voice output means that outputs a message to that effect in synthesized voice; and after the first synthesized voice output means outputs the destination name, a voice that instructs whether or not to make a call to the destination; The apparatus is equipped with a second voice recognition means that recognizes a call, and a calling means that makes a call to the destination to which the voice is output when the second voice recognition means recognizes a voice instructing to make a call.

(Function) In the telephone of the present invention, the comparison means compares the number string output from the voice recognition means with the telephone number stored in the storage means, and as a result of this comparison, the telephone number matching the number string is determined. When the telephone number is obtained, the first synthesized voice output means outputs the destination name corresponding to the telephone number in synthesized voice.

On the other hand, as a result of the comparison, if a telephone number matching the numeric string is not obtained, a message to that effect is output as synthesized voice from the second synthesized voice output means.

After the first synthesized voice output means outputs the destination name as voice, the operator confirms whether the destination name is the one desired or not, and the operator makes a voice instructing whether or not to make a call to the destination. When this voice is uttered, this voice is recognized by the second voice recognition means.

When a voice instructing a call is recognized as a result of the recognition, a call is made to the destination to which the voice is output by the calling means.

Therefore, according to the present invention, it is possible to check whether there is an error in the phone number itself input by voice or whether there is an error in voice recognition.
You can check the destination name output with synthesized voice.
As a result, it is possible to prevent calls from being sent to destinations other than the intended destination, and to significantly improve reliability.

(Example) Embodiments of the present invention will be described below based on the drawings.

FIG. 1 is a block diagram showing the configuration of a telephone according to an embodiment of the present invention.

In the telephone set shown in the figure, the numeric sounds of the telephone number uttered by the operator are input to the voice input microphone 1, and the voice signal thereof is output to the voice section detection section 2.

Here, the digit string of the telephone number is spoken one digit at a time, and "0"
Voices of numbers from 9 to 9 are input from the voice input microphone 1.

Detection of a speech section by the speech section detection unit 2 is performed by cutting out a section in which the power of the speech is equal to or greater than a threshold value set after taking into account ambient noise and the like.

The signal of the voice section extracted by the voice section detection section 2 is input to the pattern matching section 3.

The pattern matching section 3 converts the input signal into a band
Analyzes using a group of pass filters, etc., and extracts characteristic patterns of the input audio signal.

The characteristic patterns of the audio signals extracted in the pattern matching unit 3 are compared and matched with the standard audio patterns of each number stored in advance in the standard audio pattern storage unit 4, and as a result of this comparison, the feature pattern with the highest degree of similarity is High (
Alternatively, the number with the smallest distance between patterns is output as the recognition result.

As described above, the voice recognition function in the telephone is realized.

Now, the result of speech recognition is stored in the recognition result storage unit 5.
At the same time, the pattern matching section 3 informs the control section 6 of this fact.

Thereafter, the control section 6 determines that the recognition results of all the digits in one telephone number have been stored in the recognition result storage section 5, and starts the comparison section 7.

Upon receiving the activation signal from the control unit 6, the comparison unit first sends the numeric string of the telephone number stored in the recognition result storage unit 5 to the comparator 7a. The comparator 7m compares the received number string with the subscriber number pre-stored in the destination name and telephone number storage section 8 shown in FIG.

The destination name and telephone number storage section 8 stores pairs of character strings of a plurality of subscriber numbers and destination names corresponding to these subscriber numbers, each with a registration number attached thereto.

Comparison and verification in the comparator 7a is performed by matching the phone number and subscriber number, which are the recognition results, one character at a time from the beginning of each. As a result of this comparison and verification, if it is determined that there is a subscriber number that matches the phone number that is the recognition result, then the corresponding subscriber number is used! ! The green number is output to the destination name output section 7b. On the other hand, when it is determined that there is no subscriber number that matches the telephone number that is the recognition result, the registration number "0" is output to the destination name output section 7b.

If the registration number sent from the comparator 7a is other than "0", the destination name output section 7b outputs a character string representing the destination name corresponding to the registration number from the destination name and telephone number storage section 8. The character string is read and output to the control unit 6. Thereafter, a signal is sent to the control section 6 indicating that the operation of the comparison section 7 has ended.

On the other hand, the registration number "
0'' is sent, only the operation end signal of the comparison section 7 is sent from the destination name output section 7b to the control section 6.

When the control unit 6 receives the operation end signal from the comparison unit 7,
A character string indicating that the telephone number that is the recognition result does not match any subscriber number, such as "GaiTousuru Atesakiha Arimasen 4", is output to the speech synthesis unit 9. Here, the number "4" at the end indicates the position of the accent in the character string "Arimasen", that is, the position of "Se".

On the other hand, when the control unit 6 receives a character string of a destination name, for example, “Yamamoto Takashi 1” from the destination name output unit 7b, it generates a character string “Atesakiha Yamamoto Takashi Sandes 1” and sends this to the speech synthesis unit 9. Output.

When the speech synthesis unit 9 receives this character string, it converts it into synthesized speech using a predetermined regular speech synthesis method, and notifies the operator through the speaker 10 as a voice message expressing the character string.

Then, at the end of this voice message, a voice message such as "Yoroshiidesuka 3" is added to prompt the operator to confirm the destination name, and the voice message is output.

Here, the operator utters a response voice such as "yes" if the destination name of the voice message is the desired one, or "no" if it is not, so that these voices are recognized by the voice recognition function described above. The phone number obtained through voice recognition is used to make a call only when the voice ``hi'' is recognized.

Thus, according to the telephone set of this embodiment, it is possible to check whether there is an error in the input telephone number itself or whether an error has occurred in voice recognition from the destination name output in synthesized voice. It is possible to prevent calls from being sent to external destinations, and it is possible to significantly improve reliability.

In the embodiments described above, a case has been described in which a telephone number is recognized by the voice recognition function, but the present invention can be similarly applied to a telephone that makes a call using an abbreviated number instead of a telephone number.

[Effects of the Invention] As explained above, according to the telephone set of the present invention, it is possible to check whether there is an error in the input telephone number itself or whether an error has occurred in voice recognition from the destination name output in synthesized voice. As a result, it is possible to prevent transmissions to destinations other than the intended destination and improve reliability.

[Brief explanation of the drawing]

FIG. 1 is a block diagram for explaining the configuration of a telephone equipped with a voice recognition function according to an embodiment of the present invention, and FIG. 2 is a storage content of a destination name and telephone number storage unit in the telephone shown in FIG. FIG. 1... Voice input microphone, 2... Voice section detection unit,
3... Pattern matching section, 4... Standard voice pattern storage section, 5... Recognition result storage section, 6... Control section, 7... Comparison section, 7a... Comparator, 7b ... Destination name output section, 8... Destination name and telephone number storage section, 9...
-Speech synthesis section, 10...speaker. Applicant: Toshiba Corporation

Claims (1)

[Claims] (1) a storage means in which a plurality of destination names and corresponding telephone numbers are stored in advance; a first voice recognition means that recognizes voice related to the telephone number and outputs a numeric string of the telephone number; a comparison means for comparing the number string output from the voice recognition means and the telephone number stored in the storage means; and when a telephone number matching the number string is obtained as a result of the comparison by the comparison means; a first synthesized voice output means for outputting a destination name corresponding to a telephone number in a synthesized voice; and when a telephone number matching the number string is not obtained as a result of the comparison, a message to that effect is output in a synthesized voice; a second synthesized voice output means that outputs the destination name; and after the first synthesized voice output means outputs the destination name as voice,
a second voice recognition means that recognizes a voice instructing whether or not to make a call to the destination; 1. A telephone set comprising: a means for making a call to a telephone;