JP2000235697A - Voice synthesizer - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a voice synthesizer which generates again synthetic voice corresponding to words having an attribute of a specified kind by tracking back the synthetic voice stream. SOLUTION: The voice synthesizer is provided with a FIFO memory 6 to store data about a guide sentence, an analyzing part 81 to generate the guide sentence from a speaker 15 by the synthetic voice according to the data stored in the FIFO memory 6 and to detect the starting address of data about place names and numerals, a transfer width of address calculating part 3 to store the addresses of the place names and numerals in the data stored in the FIFO memory 6 and to store the first address of the data about the guide sentence which is stored last as a voice synthesis starting address P and a push switch 14 to transfer the voice synthesis staring address P to the addresses of the place names or numerals stored previously by an operation of a user.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice synthesizing apparatus, and more particularly to a voice synthesizing apparatus effective for an on-board type voice guiding apparatus such as a car navigation system.

[0002]

2. Description of the Related Art VICS for detecting the current position of a vehicle, particularly a car, and receiving and reporting traffic information and the like in the vicinity from the outside.
(Vehicle Information Communication System) has been put to practical use. Also already,
Navigation systems that provide route guidance by comparing a map database with a current position and a destination are widely used.

Further, in recent years, from the viewpoint of security, in the above-mentioned system, a configuration in which various kinds of guidance are notified by voice has been generally used.

[0004]

In the above-described conventional car navigation system that provides voice guidance, the driver is notified in real time by voice every time traffic information is generated in the system. Often miss the first part of the announcement. For example, "Route 1 and Moriguchi direction are 2
It is a traffic jam of km. It is a route that goes for a while. "
If the two guidance texts are announced by voice, the first guidance text may be missed even if the second guidance text "Route that follows the road for a while." Also, if you can hear the second half of the first guidance message, "There is a traffic jam of 2 km from Dainichi Intersection." If you miss the first half of "National Highway 1 and Moriguchi Area" Does not know whether there is congestion in the forward direction of the vehicle or in the reverse direction.

For this reason, it is conceivable to temporarily store voice guidance data in a buffer memory so that it can be generated again by operating a simple and safe means such as a push switch. All of the data stored in the buffer memory will be generated again. Buffer memory is usually FIFO (First In First Out:
Because of the first-in-first-out method, if the capacity of the buffer memory is large, the guidance that was performed before the target guidance sentence is generated again is generated. In addition, when the capacity of the buffer memory is small, there is a possibility that all of the intended guidance may not be buffered.

[0006] By the way, the information that the user really wants is limited in the voice guidance. For example, in the case of traffic congestion information of a car navigation system, the location, direction, and distance where traffic congestion occurs. Therefore, when the voice guidance is generated again, it is not always necessary to generate the voice guidance again from the beginning of the guidance text, but it is sufficient that the voice guidance can be generated from a portion desired by the user.

The present invention has been made in view of such circumstances, and it has become possible to generate a synthesized speech again by going back to a word having a specific type of attribute in a synthesized speech generated by speech synthesis. The purpose of the present invention is to provide a speech synthesizer that has been implemented.

The present invention also relates to a part necessary for a user in a sentence notified by synthetic speech, for example, a place name,
An object of the present invention is to provide a speech synthesizer suitable for a car navigation system, which can be generated again from a distance or the like.

[0009]

A speech synthesizing apparatus according to the present invention comprises a sentence data storage means for storing data of a sentence to be generated as synthesized speech, and a linguistic analysis of the data stored in the sentence data storage means. And a speech synthesizer that generates a synthesized speech by performing a language analysis on the text data stored in the text data storage means. Analysis means for detecting a start address of a word having the attribute of, and synthesized speech utterance means for uttering sentence data stored in the sentence data storage means as synthesized speech from an address designated as a speech synthesis start address. In addition to storing the start address of the word detected by the analyzing means in the data stored in the sentence data storing means, Address storage means for storing the head address of the data of the sentence as a speech synthesis start address, and the speech synthesis start address stored in the address storage means when an appropriate operation is performed. Speech synthesis start address changing means for changing to a start address of a word in the data of the sentence, wherein the synthesized speech generation means changes the speech synthesis start address by an operation of the speech synthesis start address changing means. Preferably, the sentence data stored in the sentence data storage means is generated by a synthesized speech from the changed address.

In the speech synthesizer according to the present invention, the sentence data stored in the sentence data storage means is subjected to language analysis by the analysis means, and the start address of a word having a predetermined attribute is detected. By storing the address of a word having a predetermined attribute detected by the analyzing means in the data stored in the sentence data storing means,
The head address of the last stored sentence data is stored as the speech synthesis start address. Then, by operating an address changing means such as a push switch, the speech synthesis start address stored in the address storage means is changed to the start address of a word in the data of the sentence stored before that, The sentence data stored in the sentence data storage means is generated by synthesized speech from the address after the change.

The speech synthesizing apparatus according to the present invention further comprises a map data storage means for storing map data, a fixed guidance text data storage means for storing data of a fixed guidance text relating to road traffic information, and Detecting means, and reading out the data of the fixed form guide text from the fixed form guide text data storage means in accordance with the position detected by the position detecting means in the map data stored in the map data storage means, as synthesized speech A guide sentence creating means for creating a sentence to be generated.

In such a speech synthesizer according to the present invention, the fixed form guide text data storage means stores the fixed form guide text according to the position detected by the position detecting means in the map data stored in the map data storage means. Is read and a sentence to be generated in the synthesized voice is created, so that if the user misses the voice guidance in the car navigation system, it can be easily generated again.

[0013] Further, the speech synthesis apparatus according to the present invention is characterized in that, in the above configuration, the word having the predetermined attribute is a place name and / or a number.

In the voice synthesizing apparatus according to the present invention, when a user misses voice guidance in a car navigation system, it can be easily generated again from a place name, a traffic jam distance, a distance to an intersection, and the like. Becomes possible.

[0015]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail with reference to the drawings showing the embodiments. FIG. 1 is a block diagram showing a configuration example when the speech synthesis device according to the present invention is applied to a car navigation system.

In FIG. 1, reference numeral 9 denotes a control unit of the speech synthesizer according to the present invention, and a VICS is provided outside the control unit.
Antenna 10 via tuner 11 for GPS receiver 12
, A CD-ROM drive 13, a push switch 14, and a speaker 15 are connected.

The VICS tuner 11 is in the VICS data buffer memory 2 in the control unit 9, the GPS receiver 12 is in the vehicle position detection unit 4, the CD-ROM drive 13 is in the map data buffer memory 7, The push switch 14 is connected to the control unit 9, and the speaker 15 is connected to the voice synthesis unit 8.

In the control section 9, in addition to the above-described components, a memory 1 in which fixed guide text data is stored in advance.
, An address movement width calculation unit 3 and a guide text creation unit 5.

In the memory 1, fixed guidance text data is stored in advance. Specifically, the fixed form guide text data is, for example, data such as "The road is a 2, 2, km, congested road, a route that follows a road for a while." When given to the sentence preparation unit 5, place name data is inserted before "area" and "kara".

The VICS data buffer memory 2 buffers the VICS data received by the tuner 11 and provides the buffered VICS data to the guide text creation unit 5.

The vehicle position detecting section 4 detects the vehicle position (current position) from the GPS (Groval Positioning System) signal received by the GPS receiver 12 and supplies it to the guide text creating section 5 and the buffer memory 7 for map data.

The buffer memory 7 reads out from the CD-ROM drive 13 and buffers the map data corresponding to an area within a predetermined range centered on the vehicle position detected by the vehicle position detector 4. The map data buffered in the buffer memory 7 is provided to the guide text creating unit 5.

As described above, the guide sentence creating unit 5 executes the VICS
Data is input from the data buffer memory 2, the standard guidance text data memory 1, the vehicle position detection unit 4, and the map data buffer memory 7, and a guidance text is created as described later. The created guide text is output to the voice synthesizer 8 and generated as a synthesized voice from the speaker 15 and sent to the FIFO memory 6 for buffering.

The speech synthesizing unit 8 is provided with the above-mentioned guidance text creating unit 5
Analyzing unit for language analysis of guidance text data given from
81, and a synthesized voice generating unit 82 for generating the data buffered in the FIFO memory 6 from the speaker 15 as a synthesized voice from the voice synthesis start address given from the address movement width calculating unit 3.

The analyzing section 81 of the speech synthesizing section 8 performs language analysis of the data of the guide sentence provided from the guide sentence creating section 5 to detect various information necessary for performing speech synthesis, and obtains a result of the analysis. The analysis data is stored in the FIFO memory 6. Specifically, the analysis unit 81 analyzes the part of speech information and the dependency state of each word constituting the sentence, determines the reading, accent, and pause length. The analysis result as shown in the figure is stored in the FIFO memory 6. In the figure, "S3" is a pause symbol, which is set to 100 msec in the present embodiment. “↑” and “↓” indicate accent marks for specifying whether to raise or lower the accent. "\ Plc \" indicates the start label of the place name data, and "\ n
"um \" indicates the head start label of the numerical data.

The address movement width calculation unit 3 calculates a guide text (“.”) For each unit buffered in the FIFO memory 6.
), And detects and stores the head start address of the place name data and the head start address of the numerical data in the above. Finally, the head address of the guidance sentence data stored in the FIFO memory 6 is stored as the speech synthesis start address P.

When the speech synthesis start address P stored in the address movement width calculation section 3 is given, the synthesized speech generation section 82 of the speech synthesis section 8 receives the guidance sentence data stored in the FIFO memory 6. Is synthesized from the corresponding address and output to the speaker 15, thereby generating a synthesized voice.

When the push switch 14 has been operated n times, the push switch 14 changes the speech synthesis start address P stored in the address movement width calculation unit 3 into n pieces of guide text data previously stored in the FIFO memory 6. Change to the start address of the previous place name data or numerical data. As a result, the voice synthesizer 8 reads the data buffered in the FIFO memory 6 from the voice synthesis start address P given from the address movement width calculator 3 and outputs the data to the speaker 15.

Regarding the operation of the speech synthesizing apparatus of the present invention having the above-described configuration, the guide sentence creating unit 5 provides two guide sentences, "The traffic on Route 1 and Moriguchi is 2 km from Dainichi intersection."It's a path that will take you for a while. "
The following describes an example in which a guide sentence is continuously created.

The VICS data received by the antenna 10 has revealed that a traffic jam of 2 km has occurred at the intersection ahead of the traveling direction of the vehicle equipped with the voice synthesizing device of the present invention. Suppose that it is determined that the vehicle is running in the direction of "Moriguchi" just before "Dainichi Intersection" on "National Route 1".

In this case, the guide sentence creating unit 5 reads the fixed guide sentence data stored in the memory 1 in advance, saying, "The direction is a traffic jam of 2, km from the beginning. -From the place name data "National Highway 1", "Dainichi Intersection" and "Moriguchi" read from the CD-ROM by the ROM drive 13, the guidance sentence "National Highway 1 and Moriguchi direction is 2km from Dainichi Intersection. And a navigation sentence "Route that will follow you for a while." Based on the vehicle position detected by the vehicle position detection unit 4, the control unit 9 causes the guidance text to be transferred from the guidance text creation unit 5 to the speech synthesis unit 8 at an appropriate timing. No voice is generated, and the following language analysis is performed by the analysis unit 81.

First, the analysis unit 81 analyzes the part of speech information and the dependency state of each word, determines reading, accent, and pause length, and obtains an analysis result as shown in the schematic diagram of FIG.
The data is stored in the FIFO memory 6. In FIG. 2A, "S3" is a pause symbol, which is set to 100 msec in this embodiment. “↑” and “↓” indicate accent marks for specifying whether to raise or lower the accent. “\ Plc \” indicates the head start label of the place name data, and “\ num \” indicates the head start label of the numerical data.

When the data of the analysis result obtained by the analyzing unit 81 of the speech synthesizing unit 8 is stored in the FIFO memory 6, the address moving width calculating unit 3 sends the speech synthesizing start address P, the name of each place, and the like. The start address A, B, C, D of the numerical data is stored. However, in the first guidance sentence, the address A matches the speech synthesis start address P. In the address movement width calculation unit 3, the data of the guide text is newly
Each time the information is stored in the FO memory 6, the speech synthesis start address of each guidance sentence and the head start address of each place name data and numeral data are updated and stored as described above.

As described above, from the speech synthesis start address P of the analysis data once stored in the FIFO memory 6 to the start address of the FIFO memory 6 (the last of the analysis data of the guide text stored in the FIFO memory 6 last). Is transferred to the voice synthesis unit 8 and synthesized by the synthesized voice generation unit 82 in accordance with accent symbols, pause symbols, etc., and the message "Route 1 and Moriguchi is congested 2 km from Dainichi intersection."
Is generated as a synthesized voice from the speaker 15.

Next, the second guide text is a guide text creation unit 5
Is transmitted to the speech synthesizing unit 8 and subjected to language analysis by the analysis unit 81 in the same manner as the first guide sentence described above. The result is shown in the schematic diagram of FIG.

When the analysis data of the second guidance text is transferred from the speech synthesizer 8 to the FIFO memory 6 and stored therein, the analysis data of the first guidance text previously stored in the FIFO memory 6 is used. Are moved by an address width corresponding to the data length of the entire newly stored analysis data. Therefore, the speech synthesis start address P of the analysis data of the first guidance sentence shown in FIG.
, The addresses A, B, C, and D of the first analysis data are moved by the address A ', the address width of the difference between the analysis data of the second guidance sentence and the speech synthesis start address P, respectively. It is updated and stored in the address movement width calculation unit 3 as B ', C', D '.

Then, the second analysis data is sent to the speech synthesizer 8.
The synthesized voice is synthesized by the synthesized voice generation unit 82, and is generated as a synthesized voice from the speaker 15 as "This is a route that follows the road for a while."

By the way, if the user notices that he or she missed the first guidance information at the time when the second guidance information was started to be generated from the speaker 15, the user operates the push switch 14. As a result, the control unit 9 immediately stops the speech synthesis by the synthesized speech generation unit 82 of the speech synthesis unit 8 and also transmits the speech synthesis start address P to the address movement width calculation unit 3 by the second guidance at that time. Move from the beginning of the sentence to the immediately preceding start address of the place name data or numeric data. As a result, the speech synthesis start address P moves to D 'which is the last address of the first start address of the place name data or the numerical data of the first guidance sentence, and the data after that address is read from the FIFO memory 6 by the voice. Output to the synthesizing unit 8. As a result, the guidance sentence after the address D ', that is, "2 km traffic jam. This is a route that continues for a while." Is synthesized by the synthesized voice generation unit 82 and is generated again by the speaker 15 as synthesized voice.

In the present embodiment, when the push switch 14 is operated n times, the address movement width calculator 3 moves the speech synthesis start address P to the head start address of the place name data or the numerical data of n places before. Is configured. Therefore, if the user operates the push switch 14 twice at the time when the second guidance information similar to the above starts to be generated from the speaker 15, the voice synthesis start address P
Moves to the address C 'of the first guidance message, and as a result, the guidance message "A traffic jam of 2 km from Dainichi intersection. A route that follows the road for a while."

FIG. 3 is a flowchart showing an operation procedure when the above-described speech synthesizer of the present invention is applied to a car navigation system.

It is first checked whether or not the push switch 14 has been operated (step S11), and if "NO", it is next checked whether or not a guide text has been created by the guide text creator 5 (step S12). ). If the guidance text has not been created, the process returns to step S11.

When the guide sentence is created by the guide sentence creating unit 5 ("YES" in step S12), the linguistic analysis is performed by the analysis unit 81 of the speech synthesis unit 8 (step S13), and the result is obtained. Analysis data is stored in the FIFO memory 6
(Step S14). At this time, the address moving width calculation unit 3 updates the head of the analysis data of the guidance text newly stored in the FIFO memory 6 to the speech synthesis start address P, and also updates the guidance text stored in the FIFO memory 6 until then. Each address of the data is changed according to the data length of the analysis data of the guide text newly stored in the FIFO memory 6 (step S15).

Next, the data from the voice synthesis start address P to the start address of the data stored in the FIFO memory 6 is sent to the voice synthesis unit 8 and generated by the synthesized voice generation unit 82 from the speaker 15 as synthesized voice. (Step S16).

After the data from the speech synthesis start address P to the head address in step S16 is sent to the speech synthesis unit 8 and generation of synthesized speech from the speaker 15 is started, the push switch 14 is operated. Step S11
"YES"), the voice synthesis start address P is returned to the previous address (the top start address of the place name data or the numerical data) by the number of operations n (step S21), and step S16 is performed.
Then, the generation of the guidance sentence by the synthesized voice is restarted from the address.

In this embodiment, the FIFO memory 6 has
A 128-kbit SRAM is used, and it is possible to buffer guide texts of about 8000 characters. Up to five addresses other than the speech synthesis start address P are stored in the address movement width calculation unit 3. However, it is sufficiently practical when general traffic congestion information is targeted as in the example of the guide text described above.

Further, in the present embodiment, the address held by the address movement width calculating unit 3, in other words, the address detected by the voice synthesizing unit 8 is the head address of the place name data and the numerical data, but either one is used. Is also good. Furthermore, in the present embodiment, the address held by the address movement width calculation unit 3 for application to the car navigation system is the head address of the place name data and the numerical data, but the head address of other data according to the purpose of use. May be stored in the address movement width calculation unit 3.

[0047]

As described above in detail, according to the speech synthesizer according to the present invention, the sentence data stored in the sentence data storage means is subjected to language analysis by the analysis means, and the start of a word having a predetermined attribute is started. The address is detected, and the address storage unit stores the address of a word having a predetermined attribute detected by the analysis unit in the data stored in the sentence data storage unit, and the head of the last stored sentence data. The address is stored as a speech synthesis start address. Then, by operating an address changing means such as a push switch, the speech synthesis start address stored in the address storage means is changed to the start address of a word in the data of the sentence stored before that, Since the sentence data stored in the sentence data storage means is generated by the synthesized voice from the address after the change, when the voice guidance is missed and re-generated, it can be generated without waste. .

Further, according to the speech synthesizing apparatus of the present invention, the fixed form guidance sentence data storage means stores the fixed form guidance data according to the position detected by the position detection means in the map data stored in the map data storage means. Since the sentence data is read and the guide sentence is created, it is possible to easily generate the guidance again in the car navigation system when the user misses the voice guidance.

Further, according to the voice synthesizing apparatus of the present invention, when the user misses the voice guidance in the car navigation system, the voice guidance can be regenerated from a place name, a congestion distance, a distance to an intersection, and the like. It is easily possible.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration example when a speech synthesis device according to the present invention is applied to a car navigation system.

FIG. 2 is a schematic diagram showing a result of language analysis by the speech synthesizer according to the present invention.

FIG. 3 is a flowchart showing an operation procedure when the speech synthesizer of the present invention is applied to a car navigation system.

[Explanation of symbols]

 1 Memory (for standard guide text data) 3 Address moving width calculation unit 4 Vehicle position detection unit 5 Guide text creation unit 6 FIFO memory 7 Buffer memory (for map data) 8 Voice synthesis unit 13 CD-ROM drive 14 Push switch 15 Speaker 81 Analyzer 82 Synthesized speech generator

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI theme coat ゛ (reference) G10L 13/00 G10L 3/00 Q 13/04 5/02 J F term (reference) 2F029 AA02 AB07 AB13 AC02 AC08 AC14 AC18 5D045 AB23 5H180 AA01 BB02 BB04 BB13 FF05 FF12 FF13 FF25 FF27 5J062 AA05 BB01 CC07 HH01 9A001 HH18 HH34 JJ77 KK56

Claims (3)

[Claims] 1. A sentence data storage means for storing data of a sentence to be generated as a synthesized speech, and a speech synthesis means for generating a synthesized speech by linguistically analyzing the data stored in the sentence data storage means. In the speech synthesizer provided, the speech synthesis means, by analyzing the language of the sentence data stored in the sentence data storage means, analysis means for detecting the start address of a word of a predetermined type of attribute,
And a synthesized voice uttering means for uttering the text data stored in the text data storage means as a synthesized voice from an address designated as a voice synthesis start address, wherein the text data storage means stores Address storage means for storing the start address of the word detected by the analysis means, and storing the head address of the last stored sentence data as a speech synthesis start address; and Speech synthesis start address changing means for changing a speech synthesis start address stored in the address storage means to a start address of a word in data of a sentence stored before the speech synthesis start address. When the speech synthesis start address is changed by the operation of the speech synthesis start address changing means, the text data Speech synthesis apparatus characterized by the address of the changed text data stored in the 憶 means are no order to generate a synthesized voice. 2. A map data storage means for storing map data; a fixed guidance text data storage means for storing data of a fixed guidance text relating to road traffic information; a position detecting means; Guide text creating means for reading the data of the standard guide text from the standard guide text data storage means in accordance with the position detected by the position detection means in the map data, and creating a text to be generated as synthesized speech. 2. The method according to claim 1, further comprising:
A speech synthesizer according to claim 1. 3. The speech synthesizer according to claim 2, wherein the word having the predetermined attribute is a place name and / or a number.