JPH0764763A - Information provision terminal - Google Patents

Abstract

(57) [Summary] [Purpose] To dynamically determine the output characteristic of synthetic speech according to the environment in which the user is using and to be able to output the synthetic speech with that characteristic. When a time point at which a synthetic voice output request R is input is detected by a time detection unit 1 and is notified to an output characteristic determination unit 3, a time range and a time within the range are set according to the notified time. The output characteristic storage unit 2 in which the correspondence relationship with the optimum synthetic speech output level at that time is registered is used to determine an output level commensurate with the notified time, and the output characteristic determination unit 3 outputs the output level C as the output characteristic C. Output to the synthesis output unit 4. The speech synthesis output unit 4 synthesizes the synthesized speech of the word string designated by the synthesized speech selection signal S, and outputs it as the acoustic signal A at the output level indicated by the output characteristic C from the output characteristic determination unit 3.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information providing terminal for providing information to a user by using a synthesized voice.

[0002]

2. Description of the Related Art Recently, information providing terminals that provide information to users by means of synthetic speech have become widely used because they can provide information to users in an easy-to-understand manner.
The voice synthesizing method used inside the information providing terminal of this kind is roughly classified into a recording editing method, a parameter editing method, and a rule synthesizing method described below.

First, the recording / editing method is a synthesizing method in which the voice uttered by an announcer or the like is recorded in a certain unit, and the recorded unit voice is edited to output the voice. As a method of recording voice, there is known a method of recording an analog signal on a tape recorder, a PCM for recording a digital signal through sampling, quantization, and encoding, or a method similar thereto.

Next, the parameter editing method is a method in which the voice uttered by an announcer or the like is analyzed in advance, stored in the form of a sound source and a spectrum comprehensive parameter, and synthesized and output by a synthesizer. Compared with the recording / editing method in which the waveform is stored as is, this method slightly lowers the naturalness of the synthesized speech, but can achieve a large amount of information compression. PARCOR, LSP, cepstrum analysis and the like are used as analysis / synthesis methods.

Finally, the rule synthesizing method not only simply connects the parameters obtained by the analysis, but also controls the connection method and prosody information such as pitch and amplitude according to the rules.
It combines all words and sentences.

The above-described voice synthesizing method has been conventionally used, but in an information providing terminal using such a voice synthesizing method, the characteristics of the output voice are the characteristics set at the beginning and are the same. By synthesizing words, a synthetic voice with the same characteristics (for example, the same volume and sound quality) can always be obtained.

[0007]

As described above, the conventional information providing terminal outputs the synthesized voice having the preset characteristic. Therefore, when the information providing terminal is installed in a place where the environmental noise is large, the output characteristics of the information providing terminal should be increased, for example, by increasing the output level of the synthesized voice so that the user of the information providing terminal can easily hear the synthesized voice. Need to be set.

However, since the environmental noise fluctuates, it is not always at a constant level, and when the output characteristics of the information providing terminal are set according to the state where the environmental noise is large,
If the information providing terminal is used in a state where the environmental noise is reduced, the output synthetic voice becomes noise to the surroundings. On the contrary, if the output characteristics of the information providing terminal are set according to the state where the environmental noise is small, if the information providing terminal is used in the state where the environmental noise is large, the output synthetic voice will be presented to the user. It becomes difficult to hear.

As described above, the conventional information providing terminal having the same output characteristics has a problem that it is difficult to make it easy to use in any environment. The present invention has been made in view of such circumstances, and an object of the present invention is to dynamically determine the output characteristic of synthetic speech according to the environment when the user is using the An object of the present invention is to provide an information providing terminal capable of outputting a synthesized voice having characteristics.

[0010]

The present invention focuses on the fact that environmental noise is closely related to at least one of time or date and its attribute (whether the date is a holiday or a weekday). In an information providing terminal for providing information to a user by means of both a time detection means for detecting the time when a synthetic voice output request is made and a date detection means for detecting at least one of the date and the attribute of the synthetic voice output request Means, or any one of the detecting means, and output characteristic storage means for storing the output voice of the synthetic voice corresponding to each time, each date and / or at least one of its attributes, and Based on the time of the synthetic voice output request detected by the detection means, at least one of the date and its attribute, or both, and the contents of the output characteristic storage means. An output characteristic determining means for determining an output characteristic of synthetic speech, and a voice synthesis output means for outputting synthetic speech as an acoustic signal with the synthetic speech output characteristic determined by the output determining means. Is.

[0011]

In the above structure, at least one of the time and / or the date and the attribute thereof (whether it is a holiday or the like) at the time of requesting the output of the synthetic voice, or both of them are detected, and the detected contents and the output characteristic storage means are detected. Based on the contents, the time when the synthetic voice output was requested, the date and its attributes (whether it was a holiday, etc.)
The output characteristic stored in the output characteristic storage means corresponding to at least one or both of the above is determined as the optimum output level of the synthesized voice. Then, the synthesized voice is output with the determined output characteristic.

As described above, according to the above configuration, the time or date when the information providing terminal is used and the attribute thereof are always set to the output level suitable for the environmental noise corresponding to the time or date and the attribute thereof. Since the synthetic voice is output, it is possible to always output the synthetic voice that is easily heard by the user.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a schematic configuration of an information providing terminal according to an embodiment of the present invention. The information providing terminal shown in FIG. 1 is assumed to be installed and used in a place where the level of environmental noise varies depending on the time of day.

In FIG. 1, reference numeral 1 denotes a time detection unit that receives and outputs the time at that time in response to a synthetic voice output request R from the outside, and 2 indicates the output of the synthetic voice according to (the range of) various times. An output characteristic storage unit 2 for storing characteristics in advance.

Reference numeral 3 is an output characteristic C of the synthesized voice based on the time detected by the time detection unit 1 and the contents of the output characteristic storage unit 2.
An output characteristic determination unit 4 that determines a synthesized voice output request R,
The synthetic speech selection signal S for designating (a string of words constituting the synthetic speech) to be output and the output characteristic C determined by the output characteristic determination unit 3 are received, and the synthetic speech selection signal S is received.
It is a voice synthesis output unit that synthesizes the voice designated by the method according to the output characteristic C.

Next, the outline of the operation of the information providing terminal shown in FIG. 1 will be described. First, the synthesized speech output request R and the synthesized speech selection signal S are input to the speech synthesis output unit 4, and the speech synthesis output unit 4 is requested to synthesize the speech designated by the synthesized speech selection signal S.

The synthetic voice output request R is also input to the time detection unit 1. Then, the time detection unit 1 detects the time at that time and notifies the output characteristic determination unit 3 of it. Output characteristic determination unit 3
Receives the time notification from the time detection unit 1, refers to the output characteristic storage unit 2, and outputs the output characteristic C corresponding to the notified time.
Is output to the voice synthesis output unit 4.

Then, the voice synthesis output unit 4 synthesizes the voice designated by the synthesized voice selection signal S according to the output characteristic C and outputs it as the acoustic signal A. Next, the details of the operation of each unit of the information providing terminal of FIG. 1 will be described assuming that the voice synthesis method used in the voice synthesis output unit 4 is the parameter editing method.
A case will be described as an example where a synthesized voice corresponding to the word string designated by the synthesized voice selection signal S is output.

First, the configuration of the voice synthesis output unit 4 is shown in FIG. In FIG. 2, when the synthetic voice output request R is input to the synthetic output control unit 41 and the output characteristic C determined by the output characteristic determining unit 3 of FIG. 1 is further input, the synthetic output control unit 41 At the same time as sending the parameter selection start signal PS to the parameter selection unit 42, the output characteristic signal CS specified by the output characteristic C is output to the analog amplification unit 46.

In addition to the parameter selection start signal PS from the synthesis output control unit 41, the synthesized voice selection signal S is input to the parameter selection unit 42. Parameter selection unit 42
Starts the process upon receiving the parameter selection start signal PS, and searches the combined parameter storage unit 43.

In the synthesis parameter storage unit 43, parameters (synthesis parameters) for synthesizing word voices are registered in advance for each word. There are various conceivable synthetic parameters, but here, for example, 10 ms
Is called 1 frame, and the cepstrum coefficient, pitch frequency, and voiced / unvoiced sound are measured for each frame to create a vector time series (parameter series), which is used as a synthesis parameter. In addition, as a synthesis parameter, PA
RCOR, LSP, etc. may be used.

FIG. 3 shows an example of contents stored in the synthesis parameter storage unit 43. In the example of FIG. 3, the synthesis parameter storage unit 43 has five-word synthesis parameters of “Tokyo”, “Osaka”, “weather”, “sunny”, and “rain”.
That is, a sequence in which the cepstrum coefficient, the pitch frequency, and the difference between voiced sound and unvoiced sound are measured every 10 ms is registered.

The parameter selection section 42 receives the parameter selection start signal P from the synthesis output control section 41 as described above.
The process is started according to S, and a parameter (synthesis parameter) for speech synthesis corresponding to each word (constituting a word string) designated by the synthesized speech selection signal S (for example, a word identifier) is stored as a synthesis parameter. The data is sequentially taken out from the unit 43 and output to the synthesizing unit 44 at regular intervals.

Therefore, the contents of the synthesis parameter storage unit 43 are as shown in FIG. 3, and the synthesized speech selection signal S sequentially designates the three words "of Tokyo", "weather" and "is sunny". Then, when the voice synthesis of the word string "The weather in Tokyo is sunny" consisting of these three words is requested,
The operation of the parameter selection unit 42 is as follows.

First, the parameter selection unit 42 sets the synthesis parameter of the first word "Tokyo no" to the synthesis parameter storage unit 4
Search from 3. Then, the parameter selection unit 42 reads the found synthesis parameter for one frame from the beginning and outputs it to the synthesis unit 44. Next, the parameter selection unit 42
Waits for one frame, reads the parameter for the next one frame, and outputs it to the synthesizing unit 44. The parameter selection unit 42 repeats this operation until there is no composite parameter of “Tokyo”.

Thereafter, the parameter selecting section 42 repeats the same parameter reading for the subsequent "weather" and "it is sunny", and ends the operation. By the above operation of the parameter selection unit 42, “of Tokyo”, “weather”,
This means that the parameters for synthesizing the three-word voice "It is sunny" are sequentially output to the synthesizing unit 44.

The synthesizing section 44 outputs a digital voice waveform (digital voice signal) using the synthesis parameters sequentially output from the parameter selecting section 42 in units of one frame. When the cepstrum coefficient, the pitch frequency, and the voiced / unvoiced sound discrimination are used as the synthesis parameters as described above, the synthesis unit 44 determines that the periodic sound (corresponding to the voiced sound) is generated as shown in FIG. Impulse train generator 441 for generating a pulse train of M, and an M-sequence generator 44 serving as a noise source for creating noisy speech (corresponding to unvoiced sound)
2. Switch 443 and operational amplifier 44 for switching these
4 and a logarithmic magnitude approximation (LMA) filter 445. C0, C1, ..., C in the figure
n represents the cepstrum coefficients of the 0th, 1st, ..., Nth terms, respectively. For the logarithmic amplitude approximation (LMA) filter 445, refer to “Seiji Imai, Transactions of the Institute of Electrical Communication”
80/12 Vol.J63-A No.12 ”.

The digital audio signal output from the synthesizer 44 is input to the D / A (digital / analog) converter 45. The D / A converter 45 outputs an analog audio signal by D / A converting the digital audio signal from the synthesizer 44. This analog audio signal is input to the analog amplifier 46.

The analog amplifier 46 is a D / A converter 45.
The analog audio signal (analog audio waveform) from is amplified according to the output characteristic signal CS output from the combined output control unit 41, and is output to the speaker unit 47 to output the audio.

The output characteristic signal CS output from the combined output control unit 41 (corresponding to the output characteristic C from the output characteristic determining unit 3) is, for example, the volume level of the output voice. When the output characteristic signal CS is applied, the analog amplification section 46 may change its amplification factor according to the volume level indicated by the output characteristic signal CS.

Further, it is possible to specify the power for each frequency band by the output characteristic signal CS. In this case, the sound quality can be changed. This can be realized by changing the amplification factor of the analog amplification unit 46 for each frequency band.

As described above, it is possible to realize the voice synthesis output unit 4 having the configuration shown in FIG. 2 which outputs the synthesized voice corresponding to the output characteristic C determined by the output characteristic determination unit 3. Note that the configuration of the voice synthesis output unit 4 shown in FIG. 2 is merely an example. For example, as shown in FIG. 5, the output characteristic signal CS output from the synthesis output control unit 41 is output from the analog amplification unit 46 instead of the analog amplification unit 46. It is also possible to adopt a configuration in which the D / A conversion unit 45 is provided.

In the configuration shown in FIG. 5, the D / A conversion unit 4
Reference numeral 5 denotes an operation for the input digital audio signal from the synthesizer 44 in accordance with the output characteristic signal CS (when it is desired to change the sound level, multiplication by a multiplication factor according to the output characteristic signal CS, sound quality When it is desired to change, digital filtering) is performed, D / A conversion is performed, and an analog audio signal may be output to the analog amplifier 46.

Further, as shown in FIG. 6, the output characteristic signal CS output from the combined output control section 41 may be applied to the combining section 44. In the configuration shown in FIG. 6, the synthesizing unit 44 uses the input parameter selecting unit 4
It is sufficient to output a digital audio signal with the configuration shown in FIG. 4 after performing an operation according to the output characteristic signal CS with respect to the synthesis parameter from 2 (when it is desired to change the sound quality). In addition, in order to change the sound quality, it is possible to change the pitch according to the output characteristic signal CS or change the sound level by changing the gain of the operational amplifier 444 in FIG. 4 according to the output characteristic signal CS.

Next, the details of the determination of the output characteristic by the output characteristic determining section 3 will be specifically described, including the relationship with the time detecting section 1 and the output characteristic storing section 2. First, the time detection unit 1 detects the time when the synthetic voice output request R is input and notifies the output characteristic determination unit 3 of the time. Thus, for example, when the synthetic voice output request R is input at 17:32, the time “17:32” is notified to the output characteristic determination unit 3. The output characteristic determination unit 3 refers to the output characteristic storage unit 2 based on the notified time and determines the output characteristic C of the synthetic sound to be output.

In the output characteristic storage unit 2, a time range and an output level at a time within the range are registered in advance, and an output level at a time other than the time range is registered as a default value. There is.

FIG. 7 shows an example of registration in the output characteristic storage unit 2. Here, the output level is shown in 10 steps from "0" to "9", "0" corresponds to the smallest output level, and "9" corresponds to the largest output level.
Also, "5" is used as the default value of the output level,
For the time range where the output level is not described,
"5" is used as the output level.

Therefore, in the example of FIG.
Output level is “3” until 0:00, output level is “5” from 0:00 to 7:00, output level is “7” from 7:00 to 9:00, and 9:00 to 17:00 Output level is "5" until minute, output level is "7" from 17:00 to 20:00, output level is "5" from 20:00 to 21:00, and output is from 21:00 to 24:00 Level "3" indicates that each is set.

The output characteristic determining section 3 outputs the output characteristic storage section 2 based on the time notified from the time detecting section 1 as described above.
The output level (of the synthesized voice to be output) corresponding to the time is determined with reference to
Is output to the voice synthesis output unit 4.

Therefore, as described above, the time "17: 3
In the example in which "2" is notified, the time "1:32" and the contents of the output characteristic storage unit 2 shown in FIG.
The output level in the time range to which "7:32" belongs is searched, the output level "7" is obtained, and the same level "7" is output to the speech synthesis output unit 4 as the output characteristic C.

From the above, for example, at 17:32, the synthetic voice output request R is given to the information providing terminal of FIG. 1, and the word string designated by the synthetic voice selection signal S at that time is "Tokyo", The specific operation of the information providing terminal when it is composed of the three words "weather is" and "it is sunny" is as follows.

First, the synthetic speech output request R given to the information providing terminal is inputted to the speech synthesis output section 4 and at the same time, inputted to the time detection section 1. The time detection unit 1 notifies the output characteristic determination unit 3 of the time “17:32” when the synthetic voice output request R is input.

As described above, the output characteristic determination unit 3 determines the output level "7" of the synthesized voice as the output characteristic C, and notifies the voice synthesis output unit 4 of this. Speech synthesis output unit 4
When the synthesized voice output level “7” is notified by the output characteristic C from the output characteristic determination unit 3, the synthesized voice selection signal S
Based on the words "Tokyo's", "Weather", and "It's sunny" specified by, the sound signal A "The weather in Tokyo is sunny" is output at the synthesized voice output level "7".

Similarly, for example, at 23:14, the synthetic voice output request R is given to the information providing terminal of FIG. 1, and the sequence of words designated by the synthetic voice selection signal S at that time is the same as in the previous example. When it is composed of three words "of Tokyo", "weather", and "it is sunny", the synthesized voice output level notified from the output characteristic determination unit 3 to the voice synthesis output unit 4 is "3", and " The acoustic signal A "The weather in Tokyo is sunny" is output at the synthetic voice output level "3" this time.

As described above, according to this embodiment,
It is possible to set the output characteristic of the synthetic speech to be output, and it is possible to realize an information providing terminal that outputs synthetic speech (acoustic signal A) that is always easy for the user to hear even when the level of environmental noise differs depending on the time of day.

The above description has been made on the case where the information providing terminal is installed and used at a place where the level of environmental noise varies depending on the time zone. However, the present invention is not limited to this, and for example, the date and its attribute (holiday and It is also possible to implement the present invention in an information providing terminal installed in a place where the level of environmental noise is different due to differences in weekdays.

An example of such an information providing terminal will be described below.
This will be described with reference to the block diagram of FIG. Note that FIG.
The same reference numerals are given to the same portions as. First, the information providing terminal shown in FIG. 8 includes a date detection unit 11 and an output characteristic storage unit 1.
2, an output characteristic determination unit 13, and a voice synthesis output unit 4.

The synthesized voice output request R given to the information providing terminal shown in FIG. 8 is input to the date detection unit 11 in addition to the voice synthesis output unit 4. The date detection unit 11 detects the date when the synthetic voice output request R is input and notifies the output characteristic determination unit 13 of the date. The date detection unit 11 internally has a date attribute table 111, and by referring to the table 111, the date detection unit 11 also notifies the output characteristic determination unit 13 of the attribute of the date when the synthetic voice output request R is input.

An example of the date attribute table 111 is shown in FIG. Here, the holiday is (information indicating) as the attribute of the date.
Are registered with the corresponding dates. Also, weekdays (information indicating) are prepared as the default value of the date attribute, and as the attribute of the date that is not in the date attribute table 111,
This default value will be used.

When the date attribute table 111 of FIG. 9 is used, for example, assuming that the synthetic voice output request R is input to the date detecting section 11 on March 8, the date detecting section 11 outputs the output characteristic determining section 13. Is notified of the date “03/08” and its attribute “weekday”. Similarly, if the synthetic voice output request R is input on February 11, the date "02/1
1 ”and its attribute“ holiday ”are notified from the date detection unit 11 to the output characteristic determination unit 13.

As the date, the year may also be notified. Also, the day of the week may be used as the date attribute. When the day of the week is used, this attribute can be determined without using the date attribute table (111) because this attribute is calculated from the date (year, month, day).

The output characteristic determination unit 13 is based on the date and the attribute notified from the date detection unit 11, and the output characteristic storage unit 1
2, the output characteristic C of the synthetic sound to be output is determined. In the output characteristic storage unit 12, the output level according to the date and the attribute is registered in advance.

FIG. 10 shows an example of registration in the output characteristic storage unit 12. Here, the output level is shown in 10 steps from "0" to "9", "0" corresponds to the smallest output level, and "9" corresponds to the largest output level. In FIG. 10, some fields of date and attribute have "-", which means that the item (date or attribute) of the field is not used for the search.

The output characteristic determination unit 13 refers to the output characteristic storage unit 12 based on the date and the attribute notified from the date detection unit 11 as described above, and corresponds to at least one of the date and the attribute ( The output level of the synthesized voice to be output is determined, and the output level is output to the voice synthesis output unit 4 as the output characteristic C.

Therefore, the output characteristic determining unit 13 determines, for example, that the date notified from the date detecting unit 11 is "03/0".
8 ”, the attribute is“ weekday ”, and the output characteristic storage unit 12
10 is the content shown in FIG. 10, the output level “4” corresponding to the attribute “weekday” is output to the voice synthesis output unit 4 as the output characteristic C.

Here, when the output characteristic determination unit 13 searches the output characteristic storage unit 12, the contents of the date field are prioritized over the contents of the attribute field. Therefore, for example, when the date notified from the date detection unit 11 is “07/07” and the attribute thereof is “weekday”, the output level “8” is determined as the output characteristic C.

With such a configuration, the output characteristic of the synthesized voice to be output can be set for each date and its attribute, and even if the level of environmental noise differs depending on the date and its attribute, the user can always easily listen. Synthetic voice (acoustic signal A)
It is possible to realize an information providing terminal that outputs

The above description has been made on the case where the present invention is carried out in the information providing terminal for setting the output characteristic of the synthesized voice to be output for each date and its attribute. It is also possible to set.

An example of such an information providing terminal will be described below.
This will be described with reference to the block diagram of FIG. The same parts as those in FIG. 1 or 8 are designated by the same reference numerals.
First, the information providing terminal shown in FIG. 11 includes a time detection unit 1, a date detection unit 11, an output characteristic storage unit 22, and an output characteristic determination unit 2.
3 and a voice synthesis output unit 4.

The synthesized voice output request R given to the information providing terminal shown in FIG. 11 is input to the date detection unit 11 and the time detection unit 1 in addition to the voice synthesis output unit 4. Date detector 1
1 detects the date when the synthetic voice output request R is input and notifies the output characteristic determining unit 23 of this date and its attribute. The time detection unit 1 detects the time when the synthetic voice output request R is input and notifies the output characteristic determination unit 23 of this time.

The output characteristic determination unit 23 refers to the output characteristic storage unit 22 on the basis of the date and the attribute notified from the date detection unit 11 and the time notified from the time detection unit 1, and determines the synthesized sound to be output. The output characteristic C is determined.

In the output characteristic storage unit 22, the output level according to the date, its attribute, and the time is registered in advance. An example of registration in the output characteristic storage unit 2 is shown in FIG. Here, the output level is shown in 10 steps from "0" to "9", "0" corresponds to the smallest output level, and "9" corresponds to the largest output level. 12
In some cases, the fields of date, attribute, and time (here, start time and end time) are "-". This means that the item (date, attribute or time) of that field is used for search. Indicates not to.

The output characteristic determination unit 23 refers to the output characteristic storage unit 22 based on the date and the attribute notified from the date detection unit 11 and the time notified from the time detection unit 1 as described above, Corresponding to at least one of date and attribute, time, or both (of the synthesized voice to be output)
The output level is determined, and the output level is output to the voice synthesis output unit 4 as the output characteristic C.

Therefore, the output characteristic determining section 23 determines that the date notified from the date detecting section 11 is "03/0".
8 ”, its attribute is“ weekday ”, the time notified from the output characteristic storage unit 12 is“ 18:30 ”, and the content of the output characteristic storage unit 22 is the content shown in FIG. / 08 ”attribute“ weekday ”and time“ 18:30 ”belong to the time range (17:00 to 19:59) corresponding to the output level“ 7 ”as the output characteristic C.
Output to.

Here, the priority order when the output characteristic determination unit 23 searches the output characteristic storage unit 22 is highest in the date field, and then in the order of the attribute field and the time field (time range). Therefore, for example, when the date is “07/07”, the attribute thereof is “weekday”, and the time is “08:00”, the output level “7” is determined as the output characteristic C. In addition, the date is "03/0
9 ”, its attribute is“ weekday ”, and the time is“ 02: 0
In the case of "0", the output level "3" is determined as the output characteristic C.

With such a configuration, the output characteristic of the synthesized voice to be output can be set for each date and its attribute, and further for each time, and even when the level of the environmental noise differs depending on the date, its attribute, and the time. Thus, it is possible to realize an information providing terminal that always outputs a synthetic voice (acoustic signal A) that is easily heard by the user.

In the above example, the voice synthesizing method used in the voice synthesizing output section 4 is described as the parameter editing method, but it is also possible to use the recording editing method or the rule synthesizing method.

[0068]

As described above in detail, according to the present invention,
Focusing on the fact that environmental noise is closely related to at least one of time and / or date and its attribute, the time at the time of requesting a synthesized voice and all of the date and its attribute, or only the time, or the date and its attribute. At least one of the attributes is detected, and the output characteristic registered in advance in the output storage means corresponding to the detected content is regarded as the synthetic voice output characteristic corresponding to the environmental noise when the synthetic voice output is requested. Since it is configured to output synthetic voice with characteristics,
It is possible to dynamically determine the output characteristics suitable for the environment corresponding to the date and its attribute for performing voice synthesis and the time, and output the synthesized voice that is easy to hear at any date and time used by the user. it can.

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic configuration of an information providing terminal according to an embodiment of the present invention.

FIG. 2 is a block diagram showing a first configuration example of a voice synthesis output section 4 in FIG.

FIG. 3 is a diagram showing an example of contents stored in a synthesis parameter storage unit 43 in FIG.

FIG. 4 is a diagram showing an implementation example of a combining unit 44 in FIG.

5 is a block diagram showing a second configuration example of a voice synthesis output section 4 in FIG.

6 is a block diagram showing a third configuration example of a voice synthesis output section 4 in FIG.

7 is a diagram showing an example of stored contents of an output characteristic storage unit 2 in FIG.

FIG. 8 is a schematic block configuration diagram showing another embodiment of the information providing terminal of the present invention.

9 is a diagram showing an example of a date attribute table 111 in FIG.

FIG. 10 is a diagram showing an example of stored contents of an output characteristic storage unit 12 in FIG.

FIG. 11 is a schematic block configuration diagram showing still another embodiment of the information providing terminal of the present invention.

12 is a diagram showing an example of stored contents of an output characteristic storage unit 22 in FIG.

[Explanation of symbols]

1 ... Time detection unit, 2, 12, 22 ... Output characteristic storage unit,
3, 13, 23 ... Output characteristic determination unit, 4 ... Voice synthesis output unit, 11 ... Date detection unit, 41 ... Synthesis output control unit, 42 ...
Parameter selection unit, 43 ... Synthesis parameter storage unit, 44
... synthesis section, 45 ... D / A conversion section, 46 ... analog amplification section, 47 ... speaker section, 111 ... date attribute table.

Claims (3)

[Claims] 1. An information providing terminal for providing information to a user by means of a synthetic voice, for storing time detection means for detecting a time when a synthetic voice output request is made, and storing output characteristics of the synthetic voice corresponding to each time. Output characteristic storage means, output characteristic determination means for determining the output characteristic of the synthesized speech based on the time detected by the time detection means and the stored contents of the output characteristic storage means, and the output characteristic determination means. An information providing terminal, comprising: a voice synthesizing output unit that outputs a synthetic voice as an acoustic signal with the synthesized voice output characteristic. 2. An information providing terminal for providing information to a user by means of synthetic voice, comprising: a date detecting means for detecting at least one of a date and its attribute at the time of requesting output of synthetic voice; and at least one of each date and its attribute. Output characteristic storage means for storing the output characteristic of the synthesized voice corresponding to the synthetic voice based on at least one of the date detected by the date detection means and its attribute and the stored contents of the output characteristic storage means. An information providing terminal, comprising: an output characteristic deciding means for deciding an output characteristic; and a speech synthesizing output means for outputting a synthetic speech as an acoustic signal with the synthetic speech output characteristic decided by the output deciding means. 3. An information providing terminal for providing information to a user by means of synthetic voice, comprising: time detection means for detecting the time when a synthetic voice output is requested; and at least one of the date when the synthetic voice output is requested and its attribute. Date detection means for detecting, output characteristic storage means for storing output characteristics of synthesized speech corresponding to each date and at least one of its attributes and each time, the time detected by the time detection means and the date Output characteristic determining means for determining the output characteristic of the synthesized voice based on at least one of the date and its attribute detected by the detecting means and the stored contents of the output characteristic storing means, and the synthesis determined by the output determining means. An information providing terminal, comprising: a voice synthesis output unit that outputs a synthetic voice as an acoustic signal with a voice output characteristic.