JP2000132180A - Voice output device and voice converting method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an improved voice output device capable of generating a voice more easy to listen and a method therefor. SOLUTION: A voice edition part 3 makes the telephone number received by an output data receiving part 2 into two sets of figures, and a corresponding voice data are read from a voice data base 5, thereby, the telephone number is converted into a voice data with a rising intonation of the former figures and a falling intonation of the latter figures in which the accent of the larger figure is larger in each set. A voice with the hyphen including in the data of the telephone number converted into a voice data of 'and' and a voice data of 'is' automatically added before the telephone number is generated and outputted from a voice output part 4.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an audio output device, and more particularly to an improved audio output device and an audio conversion method for generating an easily audible output audio.

[0002]

2. Description of the Related Art A voice output device used in an automatic ordering system or a reservation system by telephone outputs a number string such as a telephone number or a reservation number by voice in order to repeat input numbers or to notify a processing result. . Such a voice output device basically performs a process of converting a number to be output into voice data registered in advance and outputting the voice data. Since it is different from utterance, it is not always easy to hear the sound. Therefore, a device that generates a sound that is easy to hear has been proposed.

For example, Japanese Patent Laying-Open No. 6-59696 discloses a number reproducing method using a voice response device in which a continuous number is output as an ascending odd number and an even number and the last as a descending bottom.

[0004]

However, there is a case where unnaturalness remains by simply outputting a string of numbers by alternately changing intonation (intonation) as in the related art.

SUMMARY OF THE INVENTION The present invention has been made to solve the above problems, and an object of the present invention is to provide an improved sound output device and a sound conversion method capable of generating a sound that is easier to hear.

[0006]

In order to achieve the above object, a voice output device according to a first aspect of the present invention provides a voice output device in which voice data having different accents for each of numbers from 0 to 9 is registered in advance. A database, voice editing means for converting each data of voice output target data including a numeric string into voice data by reading out the corresponding voice data from the voice database, and synthesizing and outputting the voice data converted by the voice editing means. Voice editing means, wherein the voice editing means divides a numeric string included in the audio output target data into sets composed of two or more numbers in order from the top, and determines the magnitude relation of the numbers in each set. It determines the accent for outputting each number as a voice.

Further, the voice editing means divides the number string included in the data to be output into two consecutive sets of numbers, and increases the accent of the larger number in each set so that the accent of the larger number becomes larger. It reads out audio data.

In the voice database, voice data corresponding to each combination of accent and intonation for each number is registered, and the voice editing means determines the order of numbers in each group and the magnitude relation. It determines the accent for outputting each number as a voice.

[0009] Further, the audio editing means may include the divided two
In a set of two numbers, the voice data of each number is read so that the accent of the larger number becomes larger, the intonation of the first number rises, and the intonation of the second number falls. is there.

[0010] Further, the voice editing means has selection parameter specifying means for specifying a selection parameter indicating a characteristic tendency of the voice data read from the voice database, and the voice editing means specifies the selection parameter from the selection parameter specifying means. The voice data to be read from the voice database for a certain number is selected by the selected selection parameter.

In the voice conversion method according to the present invention, voice data having different accents for each of the numbers 0 to 9 is registered in advance, and the input voice output target numbers correspond to the numbers. In an audio output device that reads any of the audio data and outputs the audio,
The number sequence is divided into sets of two or more numbers in order from the beginning, and the voice data of each number is read out so that the accent of a large number in each group becomes large, so that the number to be output is converted to voice data. It is something to convert.

Further, in each set of two numbers, the voice data of each number is read such that the intonation of the first number becomes higher and the intonation of the second number becomes lower.

[0013]

Preferred embodiments of the present invention will be described below with reference to the drawings.

Embodiment 1 FIG. 1 is a functional block diagram showing the first embodiment of the audio output device according to the present invention. In the present embodiment, a description will be given of a voice output device 1 that converts a telephone number including a numeric string into voice data and outputs voice as an example. The audio output device 1 includes an output data receiving unit 2,
It has a voice editing unit 3 and a voice output unit 4, and further has a voice database 5 in which voice data is registered. Among them, the output data receiving unit 2 receives a telephone number sent as data to be output from a host computer or the like. The voice editing unit 3 reads out voice data corresponding to each number and characters constituting the telephone number from the voice database 5 to convert each number and the like into voice data. The audio output unit 4 synthesizes the audio data converted by the audio editing unit 3 and sends the synthesized data to the telephone network. Further, line control and the like are performed for this purpose.

FIG. 2 is a conceptual diagram showing a data configuration of the voice database 5 in the present embodiment. In the audio database 5, audio data corresponding to each combination of accent and intonation is registered for each number from 0 to 9. In other words, two accents, relatively large and small,
Four types of voice data are registered for each number by combining different types of accents, and two types of intonations, ie, rising and falling, as the intonation. More specifically, voice data with a large accent and rising intonation, voice data with a large accent and falling intonation, voice data with a small accent and rising intonation and a small accent. Then, the intonation is the voice data with the bottom falling. Further, the voice database 5 includes voice data for outputting a hyphen "-" indicating a delimiter of an area code included in telephone number data transmitted from a host computer or the like, and a telephone number of a telephone number to be voice-outputted. And speech data of words automatically added after the telephone number to indicate the end. In the present embodiment, a hyphen is output as "no" and the telephone number is added with "is" after the telephone number so that the voice is output. Therefore, each voice data is registered in the voice database 5 together with the voice data of numbers. ing. It should be noted that if the voice data other than the numbers is particularly referred to as auxiliary voice data, the four types of voice data based on the combination of the accent and intonation described above are also applied to the auxiliary voice data in the same manner as the numerical voice data. Have it ready.

A feature of the present embodiment is that the numbers constituting the telephone number are divided into two sets so that the accent of the larger number becomes larger in each set and the intonation of the preceding number is made. Is converted to voice data in which the intonation of the numeral is lowered to the bottom and the intonation of the number is lowered to the bottom, so that it is possible to generate a voice that is more audible and closer to human utterance.

Next, the operation of this embodiment will be described with reference to the flowcharts shown in FIGS. Here, a description will be given by taking “0132-4-5555” as an example as a telephone number that is sent from the host computer and outputs a voice. Note that this telephone number is used in all patterns, that is, when the first digit of a set of two consecutive numbers is larger (eg, “32”), smaller (eg, “01”), and equal. (“55”), and furthermore, when a character is processed as a set, a hyphen appears first (“−” between “2” and “4”) and a hyphen appears later (“4 This is formed for the sake of convenience to explain the processing in the present embodiment for the pattern of "-"). Of course, if we can respond to each of the above patterns,
Obviously, any sequence of numbers and hyphens can correspond to any number of digits.

When receiving the telephone number from the host computer, the output data receiving section 2 temporarily stores the telephone number in a predetermined storage area. The data of the telephone number may be in a binary format or a text format. Here, the text format is used as an example. If the data of the telephone number is stored in the buffer Str, Str [] = “0132-4-5555”, and the character data length (variable name: Strlen) is 11.

The voice editing unit 3 processes the telephone number received by the output data receiving unit 2 character by character from the beginning.
In the present embodiment, the accent and the like of each number are determined by combining two consecutive numbers as a set. The arrangement order of the characters that can constitute each set is m, and the arrangement of the characters that constitute the telephone number is m. Assuming that the order is n, m and n are initialized first (step 101). Then, the subsequent processes are sequentially repeated for each character (Str [n], n = 1 to 11) constituting the telephone number data (step 102).

First, the audio editing unit 3 takes out the first character (Str = 1) of the character (Str = 1) and determines whether or not it is a number (step 103). In this example, Str
Since [1] is 0 and is the first number (m = 1) in the set, the number 0 is temporarily stored in the buffer Number1 (steps 104 and 105). Then, the processing target is shifted to the next character by incrementing n, and m = 2 is set in order to make the next character a pair with the number stored in Number1 (step 106).

The voice editing unit 3 outputs the next (n = 2) character (S
tr [2]) is taken out and it is determined whether or not it is a number (step 103). In this example, Str [2] is 1, and since it is the second (m = 2) number in the set, the number 1 is temporarily stored in the buffer Number2 (steps 104 and 107). Here, the audio data conversion processing shown in FIG. 4 is performed (step 108).

In FIG. 4, here, the character Str [n] to be processed is a number, and the voice data conversion processing is performed when a pair can be formed by two numbers.
Move on to In step 203, Num constituting the set
Compare ber1 and Number2. Here, N
Since the number 1 (= 0) is smaller than the number 2 (= 1), the voice data of the number 0 whose accent is small and the intonation is ascended with respect to the first number 0 temporarily stored in the number 1 is read out from the voice database 5 and output. Predetermined storage area An for storing audio data
The data is stored at the head storage position of n (Ann [(n-1) = 1]) (step 204). In addition, the audio editing unit 3
The second digit 1 temporarily stored in mber2 is read out of the audio database 5 with the accent 1 being large and the intonation falling down from the audio database 5 and the next storage position (Ann [n = 2] in the storage area Ann ) (Step 205).

As described above, the telephone number 1
When the conversion of the set of numbers into audio data is completed, n
Is incremented, the processing target is moved to the next character, and m is initialized (step 109). It is to be noted that Num for temporarily storing numbers to be processed in the audio data conversion is used.
It is desirable that the contents of ber1 and Number2 be also initialized at this point for the subsequent processing.

The voice editing unit 3 outputs the next (n = 3) character (S
tr [3]) is taken out, and it is determined whether or not it is a number (step 103). In this example, Str [3] is 3, and it is the first number (m = 1) in the set. Therefore, the number 3 is stored in the buffer Number1 in the same manner as described above.
Is temporarily stored, and n is incremented to m = 2 (steps 104 to 106).

Subsequently, the voice editing unit 3 extracts the next (n = 4) character (Str [4]) and determines whether or not it is a number (step 103). In this example, Str
[4] is 2, and the second (m =
2) The number 2 is stored in the buffer Number
Then, the audio data conversion processing is performed (Steps 104, 107 and 108).

In FIG. 4, the processing is performed in substantially the same manner as described above, but Number 1 (= 3) constituting this group is
Since it is larger than Number2 (= 2), Number
The voice data of the number 3 whose accent is large and the intonation rises from the first number 3 temporarily stored in 1 is read out from the voice database 5 and stored in the storage area Ann.
Is stored in the next storage position (Ann [(n-1) = 3]) (step 206). The audio editing unit 3 also stores the second number 2 temporarily stored in Number 2 in this group.
Then, the voice data of the numeral 2 with a small accent and a lower intonation is read out from the voice database 5 and stored in the next storage position (Ann [n = 4]) (step 207).

As described above, the telephone number 1
When the conversion of the set of numbers into audio data is completed, n
Is incremented, the processing target is moved to the next character, and m is initialized (step 109).

The voice editing unit 3 outputs the next (n = 5) character (S
tr [5]) is taken out, and it is determined whether or not it is a number (step 103). Since a hyphen is used in this example, audio data conversion processing is performed (step 110).

In FIG. 4, the process proceeds to step 208 because the character Str [n] to be processed is not a numeral.
In step 208, here, when m = 1, that is, when there is no number temporarily stored in Number1 to form a set, the accent is small for Str [5], and the intonation is downward. The voice data of the hyphen is read from the voice database 5 and is stored at the next storage position (Ann [n =
5]) (step 210). That is, in response to the fact that the hyphen inserted at the boundary of the area code or the like of the telephone number is normally uttered as "no", the audio editing unit 3 in the present embodiment sets the hyphen to the voice "no". Converted to data. Then, the processing target is moved to the next character by incrementing n, and m is initialized (step 111). Here, since m should be 1, the initialization of m may be omitted.

Subsequently, the voice editing unit 3 takes out the next (n = 6) character (Str [6]) and determines whether or not it is a number (step 103). In this example, Str [6]
Is 4 and is the first number (m = 1) in the set, so the number 4 is temporarily stored in the buffer Number1 (steps 104 and 105). Then, the processing target is shifted to the next character by incrementing n, and m = 2 is set in order to make the next character a pair with the number stored in Number1 (step 106).

The voice editing unit 3 outputs the next (n = 7) character (S
tr [7]) is taken out and it is determined whether or not it is a number (step 103). Since a hyphen is used in this example, audio data conversion processing is performed (step 110).

In FIG. 4, the process proceeds to step 208 because the character Str [n] to be processed is not a numeral.
In step 208, here, when m = 2, that is, when the number 4 is temporarily stored in Number 1 to form a group, the accent of the number 4 is large, and the intonation of the number 4 is increased. Data is read from the voice database 5 and stored in the storage area A
nn is stored in the next storage position (Ann [(n-1) = 6]) (step 209). Then, the voice data of the hyphen with the small accent and the lower intonation of the hyphen is read from the voice database 5 and stored in the subsequent storage position (Ann [n = 7]) (step 210). In other words, the number temporarily stored in Number1 is located at the beginning when a set that could not form a set can be formed, so that the accent is large.
The intonation is converted to ascending audio data. A hyphen is processed as a single character, and is always converted into voice data with a small accent and a lower intonation without forming a set. Then, n
Is incremented, the processing target is moved to the next character, and m is initialized (step 111).

The voice editing unit 3 outputs the next (n = 8) character (S
tr [8]) is taken out and it is determined whether or not it is a number (step 103). In this example, Str [8] is 5 and is the first number (m = 1) in the group, so the number 5 is temporarily stored in the buffer Number1 (steps 104 and 105). Then, by incrementing n, the processing target is shifted to the next character and the next character can be paired with the number of Number1, so that m = 2
(Step 106).

Subsequently, the voice editing unit 3 extracts the next (n = 9) character (Str [9]) and determines whether or not it is a digit (step 103). In this example, Str
[9] is 5, and the second (m =
2) The number 5 is stored in the buffer Number
Then, the audio data conversion processing is performed (Steps 104, 107 and 108).

At step 203 in FIG. 4, Number1 (= 5) constituting this group is replaced with Number2.
(= 5), the voice data of the number 5 whose accent is large and the intonation rises from the first number 5 temporarily stored in the number 1 is read out from the voice database 5, and the next storage position of the storage area Ann ( A
nn [(n-1) = 8]) (step 20).
6). In addition, the audio editing unit 3 determines that Numb
The second digit 5 temporarily stored in er2 is read out from the voice database 5 with the voice data of the digit 5 whose accent is small and the intonation is lower, and is stored in the subsequent storage position (Ann [n = 9]). (Step 20
7). In the present embodiment, Number1 and Nu1
If the numbers stored in mber2 are equal, Number
Although the case where r1 is larger than Number2 is treated as the same, the case where r1 is smaller may be treated as the same. Alternatively, a separate conversion rule for audio data may be set.

As described above, the telephone number 1
When the conversion of the set of numbers into audio data is completed, n
Is incremented, the processing target is moved to the next character, and m is initialized (step 109).

The voice editing unit 3 takes out the next (n = 10) character (Str [10]) and determines whether or not it is a digit. However, Str [10] = Str [11] = 5. Since the process is the same as the above-described process of the relationship of Str [8] = Str [9] = 5, the description is omitted.

Further, the voice editing unit 3 attempts to retrieve the next (n = 12) character (Str [12]), but since it does not exist, performs voice data conversion processing for termination (step 112).

In FIG. 4, here, the end (n> St)
rlen), the process proceeds to step 211. In step 211, here, when m = 1, that is, when the number is not temporarily stored in Number1, the voice data indicating that the accent is small and the intonation ends in the bottom is read out from the voice database 5, and the subsequent Storage location (Ann [n = 1
2]) (step 212). In the present embodiment, “is” is registered as audio data corresponding to this end. Also, if m = 2 in step 211, the number is temporarily stored in Number1, so before registering the audio data indicating the end,
The voice data of the number whose accent is large and intonation rises is read out from the voice database 5,
The data is stored at the storage position (Ann [(n-1)]) immediately before the end (step 212).

As described above, the voice editing unit 3 converts the telephone number into the corresponding voice data and stores it in the storage area Ann. Thereafter, the audio output unit 4 outputs the contents of Ann as audio via the telephone network. The voice to be announced is “5555 of 0132”. FIG. 5 shows the magnitude of the accent of the read audio data and the rise / fall of the intonation.

In the present embodiment, when a pair can be formed by two consecutive numbers in the numeral string of the telephone number to be output as a voice, the two numbers in the pair are compared and the two numbers which are equal or larger are compared. Speech conversion is performed so as to increase the accent and decrease the accent of the smaller number. This makes it possible to make the output sound smoother and easier to hear. Furthermore, in each group, the intonation of the voice data of the preceding number is raised,
By making the intonation of the audio data of the subsequent numeral lower, it is possible to make the output audio more audible.

Embodiment 2 FIG. 6 is a functional block configuration diagram showing Embodiment 2 of the audio output device according to the present invention. The audio output device 1 according to the present embodiment includes a selection parameter specifying unit 6 in addition to the configuration described in the first embodiment. The audio editing unit 3 reads out from the audio database 5 audio data corresponding to the character to be processed according to the type of the character before and after the character and the magnitude of the number. Gives a selection parameter to the audio editing unit 3 as an index when reading audio data from the audio database 5. With such a configuration, the following effects can be obtained.

For example, in Japan, there are regions where conversations are made with accents and intonations slightly different from standard ones such as raising the tail of the sentence. In the first embodiment, each character is uniformly converted to voice data in accordance with the size of numbers and the arrangement of characters irrespective of such local intonation. Therefore, the present embodiment is characterized in that a character to be output can be output as a voice with a non-standard accent and intonation, so that an announcement can be made with a voice that is more familiar in each region. That is, for example, if the selection parameters 1 are assigned in accordance with the characteristic tendency of the intonation, such as 1 in the Hokkaido region and 2 in the Kansai region, the audio output device 1 according to the present embodiment can be specified. If used only in a local area, the announcement can be made in a very familiar and audible sound in that specific local area. In the first embodiment, the auxiliary voice data other than the lower accent is used because the accent is small and the intonation is not reduced. In the voice database 5 used in the first embodiment, four types of auxiliary voice data are prepared similarly to the voice data of numbers.

Next, the operation of the present embodiment will be described. However, only a part of the audio data conversion processing in the audio editing unit 3 is different from that of the first embodiment, and only the processing of that part will be described. . FIG. 7 shows a process added to the audio data conversion process shown in FIG.

For example, when the conversion of the telephone number into the voice data is completed and “is” indicating the end is added to the voice data, the voice editing unit 3 sets the accent to be small, intonation depending on the value of the designated selection parameter. Is read from the voice database 5 (steps 213-1 and 213-3) or voice data with a large accent and a rising intonation (step 213-2).

As described above, the selection parameter can be designated as an index for reading any audio data having different accents or the like. Therefore, even if the same telephone number is used, the telephone number can be changed with different accents depending on the announcement destination. Can output voice. The selection parameters are
What is necessary is just to determine by the system or operation which uses the audio output device 1, such as fixing by the installation location of the audio output device 1 or changing dynamically by the area code of the telephone number of a line connection destination. Although FIG. 7 illustrates the case where the value of the selection parameter is three, the number of branches is not limited to this. In addition, since the audio data can be read in accordance with the value of the selected parameter, it is quite possible that the same audio data is read with different parameter values as in the example of FIG.

FIG. 7 shows an example in which the process of step 213 is divided by the value of the selection parameter as a representative, but the selection parameter is one condition for which audio data is read from the audio database 5. Therefore, steps 204 to 207 in FIG.
In 209, 210, and 212, the same branch processing as in step 214 is added.

In the present embodiment, the selection parameter designating section 6 provided as the selection parameter designating means is
Although the selection parameter is assigned to the selection parameter, the selection parameter designation unit may be a simple selection parameter storage unit. Further, in the present embodiment, any one of the voice data is read out from the single voice database 5, but for example, a different voice database 5 is prepared for each region, and the voice database 5 is switched.
Alternatively, the same effect can be obtained even if the installed voice database 5 is replaced.

In each of the above embodiments, two types of accents, a relatively large accent and a small accent, are registered in advance in the speech database 5 as different accents. , Three or more accents may be prepared. In each of the above-described embodiments, two numbers are grouped in the order of arrangement to determine the magnitude of the accent of each number. If it is a character string of about 10 digits such as a telephone number,
Although a set of two numbers is sufficient in terms of rhythm, a case in which a number having a large number of digits is announced or a case where it is better to consider accents in a set of three or more numbers may be considered. Thus, the case where three or more numbers are grouped is also within the scope of the present invention.

Also, the case where the present invention is applied to voice output of telephone numbers has been described as an example. However, the present invention can also be applied to a system that announces voice target data including other digit strings, such as card numbers and reservation numbers, by voice. Needless to say.

[0051]

According to the present invention, the magnitude of the accent is determined based on the magnitude relation of the numbers in each set obtained by dividing the number sequence. In particular, since the voice data is read so that the accent of the larger number becomes larger in each group, it is possible to generate and output a more audible and less unnatural voice.

Further, in addition to the magnitude of the accent, each set of numbers is converted into voice data of a rising or falling intonation depending on the arrangement.
A more audible sound can be generated and output.

Further, since it is possible to select audio data to be read from the audio database when converting into audio data, it is possible to generate audio other than the standard, for example, adapted to local unique accents and intonations. As a result, it is possible to output a sound that is more familiar in each region.

[Brief description of the drawings]

FIG. 1 is a functional block configuration diagram showing a first embodiment of an audio output device according to the present invention.

FIG. 2 is a conceptual diagram showing a data configuration of a voice database 5 according to the first embodiment.

FIG. 3 is a flowchart showing a process of converting to audio data in the first embodiment.

FIG. 4 is a flowchart showing a voice data conversion process in FIG. 3;

FIG. 5 is a diagram illustrating an example of a processing result according to the first embodiment;

FIG. 6 is a functional block diagram showing a second embodiment of the audio output device according to the present invention.

FIG. 7 is a flowchart showing only a part of audio data conversion processing according to the second embodiment.

[Explanation of symbols]

1 audio output device, 2 output data receiving unit, 3 audio editing unit, 4 audio output unit, 5 audio database, 6 selection parameter specifying unit.

Claims (7)

[Claims] 1. A voice database in which voice data having different accents for each of numbers from 0 to 9 are registered in advance, and voice output target data including a digit string by reading out the corresponding voice data from the voice database. And audio output means for synthesizing and outputting the audio data converted by the audio editing means.The audio editing means converts a numeric string included in the audio output target data into An audio output device, which is divided into sets each composed of two or more numbers in order from the top, and determines an accent for outputting each number as a voice according to the magnitude relation of the numbers in each set. 2. The voice editing means divides a numeric string included in the audio output target data into a set of two consecutive numbers, and adds a larger number to each set such that the accent of a larger number is larger in each set. 2. The audio output device according to claim 1, wherein the audio data is read. 3. The voice database has registered therein voice data corresponding to each combination of accent and intonation for each number, and the voice editing means determines the order of the numbers in each group and the magnitude relation. 2. The voice output device according to claim 1, wherein an accent for outputting each number by voice is determined. 4. The voice editing means is arranged such that, in a set consisting of two divided numbers, the accent of the larger number is larger, the intonation of the first number is ascended, and the intonation of the later number is ascended. 4. The audio output device according to claim 3, wherein the audio data of each of the numbers that fall is read. 5. The apparatus according to claim 1, further comprising: a selection parameter designating unit for designating a selection parameter indicating a characteristic tendency of the audio data read from the audio database by the audio editing unit. 4. The voice output device according to claim 1, wherein voice data to be read from the voice database for a certain number is selected according to the selected selection parameter. 6. Voice data with different accents for each of the numbers 0 to 9 is registered in advance, and one of the voice data corresponding to the number is read out for the input voice output target number. In a voice output device for outputting voices, a digit string is divided into sets of two or more numbers in order from the beginning, and the voice data of each number is read out so that the accent of a large number in each set becomes large. A voice conversion method, comprising converting a voice output target number into voice data. 7. The method according to claim 1, wherein in each set of two numbers, voice data of each of the numbers is read such that the intonation of a further number rises upward and the intonation of a subsequent number falls downward. 6. The voice conversion method according to 6.