JPH09281988A - Voice synthesizer - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make traffic information easy to hear by retrieving to which fundamental attribute pattern the attribution of an extracted word string belongs, and inserting words into insertion patterns based on the retrieval result. SOLUTION: The language analyzing part 51 of a voice synthesizing part 5 inputs character broadcasting data from an FM receiver to generates KANJIs (Chinese characters), characters. etc., in a word string in which word attributions are mixed. Moreover, a word dictionary 52 includes information of attributions indicating kinds of words and applies the information to the language analyzing part 51. Then, an insertion part 53 insertingly synthesizes the word string generated by the language analyzing part 51 in an easy-to hear form. Further, a voice series making part 54 makes the series of an insertingly synthesized sentence in voice signs. Furthermore, a rythm imparting part 55 imparts Japanese reading syllabaries, accent information and information of pause and intonation on the voice signs. Mainly, words of a KANJI sentence constituting traffic information are inserted into insertion patterns consisting of a postpositional word functioning as an auxiliary to a main word. a conjunction, a verb or the like and then the words become a text which is easy to hear as voice information.

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 voice synthesizing a character string mainly composed of Chinese characters obtained by receiving FM multiplex character broadcasting with high sound quality.

[0002]

2. Description of the Related Art Obtained from the reception of the above FM multiple character broadcasting,
Traffic information consisting mainly of Chinese characters is used for display. It is more convenient if the traffic information obtained from the receiver mounted on the vehicle is output as a display and at the same time as a voice. This is because the driver should not look away from the front as much as possible for safety.

As a device for synthesizing the traffic information into voice, there is a text-to-speech synthesizer capable of using a kanji / kana mixed sentence of the traffic information as an input text.

[0004]

However, the traffic information obtained mainly from the Chinese character sentence obtained from the FM multiple character broadcasting is made for display, and the contents are obvious at a glance. However, even if the voice is used as it is, particles, conjunctions, verbs, etc. are omitted, and there is a problem that it is difficult to hear as a spoken word.

Therefore, in view of the above problems, it is an object of the present invention to provide a voice synthesizing device which makes it easy to hear traffic information mainly composed of Chinese characters obtained from FM multiple character broadcasting.

[0006]

In order to solve the above-mentioned problems, the present invention provides an FM multiple character in a voice synthesizing apparatus for linguistically processing a text, performing a phonetic symbol sequence, and adding a prosody to perform voice synthesis. Words of Chinese characters and symbols that mainly constitute traffic information obtained from broadcasting, a word dictionary that stores attribute information indicating the type of each word, and a word string with attributes by performing language processing using the traffic information as text. Which has a plurality of predetermined basic attribute patterns, and an embedding pattern that grammatically connects the basic attributes of the basic attribute patterns, and which basic attribute is the attribute of the extracted word string. It is provided with a fitting and synthesizing unit which searches for whether the character belongs to a pattern and fits a word based on the search result. By this means, the words of the kanji sentence that make up the traffic information are embedded in the embedding pattern consisting of particles, conjunctions, verbs, etc., and the sentence becomes easy to hear as voice information.

The embedding / synthesizing unit defines the two words connected between the symbols as the adjectives of the preceding word and the preceding word.
Convert to one word. By this means, it is possible to embed the FM multiplex character broadcast corresponding to various expressions. A phonetic symbol sequence is applied to the converted words to add prosody. By this means, it becomes possible to give an optimum prosody in correspondence with various expressions.

In the search of the inset synthesizing unit, even if the attribute of the extracted word string does not match the basic attribute pattern, if there is the closest one, it is attributed to it. By this means, it is possible to widen the application range of the fitting. When the order of the extracted attributes of the word string is different from the arrangement order of the basic attribute pattern, it is considered that the attributes of the extracted word string are rearranged and belong to this in the search by the inset combining unit. This measure likewise makes it possible to widen the application range of the fitting.

[0009]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a diagram for explaining a speech synthesizer according to the present invention. In the figure, an antenna 1 for receiving FM multiplex character broadcasting, an FM receiver 2 for demodulating a received signal, a speaker 3 for outputting FM broadcast sound of the demodulated signal, and a Chinese character sentence of traffic information of the demodulated signal. , A voice synthesizer 5 for synthesizing voice to make it easy to hear Chinese characters and the like of traffic information, and a synthetic sound from the voice synthesizer 5 and a broadcast sound from the FM receiver as alternatives. And a switch 6 for selecting.

FIG. 2 is a diagram showing a schematic configuration of the voice synthesizing section 5 of FIG. As shown in the figure, the voice synthesis unit 5 is
A language analysis unit 51 that inputs teletext data from a receiver and generates a Chinese character sentence or the like into a word string in which word attributes are mixed, and a word that includes attribute information indicating the type of word and provides this information to the language analysis unit 51. A dictionary 52, an inset synthesizing unit 53 for performing inset synthesizing of the word string generated by the language analyzing unit 51, which will be easy to hear, as will be described later, and a speech sequencer 54 for sequencing the inset-combined sentences into phonetic symbols. , And a prosody providing section 55 for providing information on phonetic alphabets, accent information, poses, and intonations.

FIG. 3 is a view for explaining an example of attributes given to words in the word dictionary 52 of FIG. As shown in the figure, an attribute A is given to a word representing a unique place name, for example, Ihagi A, Ishikawa Town A, Bay Bridge A, and Haneda A. An attribute B is given to a word representing a general point, for example, level crossing B. An attribute C is given to a word representing a unique route name, such as Yokohane C and Ringhachi C. An attribute D is given to a word representing a general road, for example, a line D. An attribute H is given to a word representing time, for example, 21 minutes H. The attribute I is given to the word representing the distance, for example, 1.5 KI. The words that represent directions include, for example, upper F and inner F,
An attribute F is given (upper means “upward”, and inner means “inner turn”). An attribute K is given to the symbol representing →, for example, → K. The attribute Z is given to the symbol indicating (), for example, (Z,) Z.

[0012] Generally, attributes are assigned to the word dictionary 52 so as to correspond to the types of words included in the FM multiplex character broadcast. FIG. 4 is a flow chart for explaining the processing of the language analysis unit 51 and the embedded synthesis unit 53 of FIG. 2, and FIG. 5 is a diagram showing a first example of the processing of FIG.
FIG. 8 is a diagram showing a second example of the processing of FIG. 4.

In step S1, word attribute assignment (see FIGS. 5B and 6B) is given to the teletext data example (see FIGS. 5A and 6A) input to the language analysis unit 51. Is done. An example of these teletext data is the display 4
Is displayed in. In step S2, the teletext data is cut with the attributes of K and Z to form a new word string (see FIG. 5).
(C) and FIG. 6 (c)).

In step S3, "Ihagi A level crossing B" is converted into an attribute of "Ihagi level crossing B" within the range of the new word subjected to the cutting process, as shown in FIG. 6 (d). This is because the point “railroad crossing B” may be described by the place name “Ihagi A”. Along with such conversion, the unique route name “ring eight C” is converted to the road “ring eight D”. Similarly, FIG.
As shown in (d), the route name “Yokohane C” is converted to the road “Yokohane D” and the place name “Ishikawacho A” is converted to the point “Ishikawacho B”. When "route L" is described by the place name "Baybridge A", "Baybridge A route L" is converted to "Baybridge L".

When the road "line D" is described by the route name "yokowa" such as "yokohan C line D", it is converted to "yokohan line D". In step S4, the word string in which the new attribute is converted is cut into words for each attribute (see FIG. 5 (e) and FIG. 6 (e)). In step S5,
Teletext data is represented by an example sentence pattern of attributes (Fig. 5
(D), FIG.6 (d)).

In step S5, a sentence example pattern is searched. FIG. 7 is a flowchart illustrating the sentence example pattern search processing in step S5 of FIG. Step S
In 11, it is confirmed that the extraction of the sentence example pattern is completed. In step S12, the basic attribute patterns up to the sentence examples 01, 02, ... The example sentence pattern in FIG.
The sentence example pattern of FIG. 6 is closest to the pattern of sentence example 01, and matches the pattern of sentence example 02.

In step S13, the words of the sentence example pattern are rearranged in accordance with the basic attribute pattern. In step S6, the following embedded composite data generation is performed. In order to make the sentence easy to understand as voice information, the embedded example synthesizer 53 preliminarily prepares sentence examples 01, 02, ... nn.
On the other hand, a fitting pattern composed of particles, connections, verbs, etc. is stored, and each rearranged word is fitted into this fitting pattern as follows.

[0018] In the example of FIG. 5, the horizontal wings on rehearsal, via the Bay Bridge from Ishikawa-cho to <br/> Haneda it will take 21 minutes.
In the example of (underlined inset pattern) 6, the inner around the ring bees, 1.5K from Ihagi crossing
In m, outside around is the 1.5K m to Ihagi railway crossing.

The inset sentence thus obtained is output as voice. This makes it easier to hear,
The traffic information of the FM text multiplex broadcast is formed into a high-quality synthesized sound.

[Brief description of drawings]

FIG. 1 is a diagram illustrating a speech synthesizer according to the present invention.

FIG. 2 is a diagram showing a schematic configuration of a voice synthesis unit 5 in FIG.

FIG. 3 is a diagram illustrating an example of attributes given to words in the word dictionary 52 of FIG.

FIG. 4 is a language analysis unit 51 and a fitting synthesis unit 53 of FIG.
5 is a flowchart illustrating the processing of step S1.

5 is a diagram showing a first example of the process of FIG.

FIG. 6 is a diagram showing a second example of the processing of FIG.

FIG. 7 is a flowchart illustrating a sentence example pattern search process in step S5 of FIG. 6;

[Explanation of symbols]

 5 ... Speech synthesis unit 52 ... Word dictionary 51 ... Language analysis unit 53 ... Inset synthesis unit

Claims (5)

[Claims] 1. A speech synthesizer for subjecting text to language processing, performing a phonetic symbol sequence, and adding a prosody to speech synthesis, including kanji and symbol words that mainly constitute traffic information obtained from FM multiple character broadcasting, A word dictionary that stores attribute information indicating the type of each word, a language processing unit that performs language processing using the traffic information as text, and extracts a word string with attributes, and a plurality of predetermined basic attribute patterns And a fitting pattern that grammatically connects the basic attributes of this basic attribute pattern, searches which basic attribute pattern the attribute of the extracted word string belongs to, and sets the word fitting pattern based on the search result. A voice synthesizing device having an embedded synthesizing unit to be embedded therein. 2. The embedding / synthesizing unit converts two words connected between the symbols into one word by using a preceding word as an adjective of a succeeding word.
The speech synthesizer according to. 3. The speech synthesizing apparatus according to claim 2, wherein a phonetic symbol sequence is performed on the converted words to add prosody. 4. In the search by the inset synthesizing unit, even if the attribute of the extracted word string does not match the basic attribute pattern, if there is a closest one, it is considered to belong to this. The speech synthesizer according to claim 1. 5. When the attribute of the extracted word string is different from the arrangement order of the basic attribute pattern in the search by the embedding / synthesizing unit, the attribute is rearranged and regarded as belonging to it. The speech synthesis apparatus according to claim 4, wherein: