JP2758851B2 - Automatic translation device and automatic translation device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION The present invention further comprises an automatic translator for translating a character string composed of an input utterance voice sentence in a source language into a character string in a target language, as well as voice recognition means and voice synthesis means. It relates to an automatic interpreter.

[0002]

2. Description of the Related Art
When a sentence is generated by a language translation, a dialogue system, or the like, processing is generally performed in units of a sentence (for example, Tsuyoshi Morimoto, “Toward Realization of Automatic Translation Telephone”, Information Processing, V
ol. 35, No. 1, January 1994). However, when an input is given in chronological order as in an automatic interpreter or the like, in the processing for each sentence, the processing cannot be started until the utterance is completed, and the transmission of the utterance is delayed. As a result, there were the following problems. (1) Unity between utterances is broken. (2) The processing time increases. The reason that the user uses the automatic interpreting apparatus is that it is possible to achieve his or her own purpose with the same burden as performing a normal conversation. Accordingly, it is very important to solve the above-mentioned problem that places a great burden on the user other than the intended purpose for constructing a practical automatic interpreting apparatus.

[0003] An object of the present invention is to solve the above problems, to perform translation in a shorter time than in the conventional example, and to translate the sentence into a natural sentence so that the whole sentence is meaningful. A translation device and an automatic translation device are provided.

[0004]

According to the present invention, there is provided an automatic translation apparatus for translating a character string composed of an input utterance voice sentence in a source language into a character string in a target language. A storage device that stores a plurality of division patterns in the source language utterance voice sentence in advance, and a character string composed of the input source language utterance voice sentence, with reference to the plurality of division patterns stored in the storage device, An original sentence dividing unit for dividing the character string, a partial translating unit for translating the character string divided by the original sentence dividing unit into a character string of a translation result of a target language for each divided unit; All translation means for translating into a character string of a translation result of a target language without dividing a character string composed of uttered speech sentences of the source language, and a first translation string which is a current translation result by the all translation means. String and all of the above The first character string is compared with the second character string obtained by concatenating the character string of the current translation result by the partial translation means after the character string of the previous translation result by the translation means. If the second character string matches and the word order is the same, the character string of the current translation result by the partial translation means is output as the translation result of the target language, and then the first character string and the partial The character string of the current translation result by the translation means is compared with the third character string obtained by concatenating the character string of the previous translation result by the all translation means, and the first character string is compared with the third character string. When the third character string matches and the word order is reversed, the translation result obtained by omitting redundant expressions from the current translation results by the all translation means is output as the translation result of the target language, while the first translation result is output. Character string and the third
And a translation generating means for outputting the current translation result by the above-mentioned all translation means as a translation result of the target language when the character strings do not match and the word order cannot be matched.

According to a second aspect of the present invention, there is provided an automatic translating apparatus, wherein the automatic translating apparatus recognizes the uttered voice based on an input voice signal of the source language. Thus, the apparatus further comprises a voice recognition unit that converts the character string into the source language and outputs the converted character string to the original sentence division unit.

According to a third aspect of the present invention, there is provided an automatic interpreting apparatus for converting a character string of a translation result of a target language output from the translation generating means into a voice signal. And a speech synthesizing means for outputting the speech.

[0007]

In the automatic translator according to the first aspect of the present invention, the storage device stores in advance a plurality of divided patterns in the utterance voice sentence in the source language. Then, the original sentence dividing means divides the input character string including the uttered voice sentence in the source language with reference to a plurality of division patterns stored in the storage device. Next, the partial translation unit translates the character string divided by the original sentence division unit into a character string of a translation result of a target language for each divided unit, while the full translation unit inputs the character string. The translation is performed on the character string of the translation result of the target language without dividing the character string composed of the uttered voice sentence of the source language. Further, the translation generating means includes a first character string which is a character string of a current translation result by the all translation means, and a partial translation means after a character string of a previous translation result by the all translation means. Is compared with a second character string obtained by concatenating the character strings of the current translation result according to the above. If the first character string matches the second character string and the word order is the same, the partial translation is performed. Outputting the character string of the current translation result by the means as a translation result of the target language; and then adding the first character string and the character string of the current translation result by the partial translation means to the first translation by the full translation means. A third character string obtained by concatenating the character strings of the immediately preceding translation result is compared. If the first character string matches the third character string and the word order is reversed, The translation result obtained by omitting redundant expressions from the current translation results If the first character string and the third character string do not match and the word order cannot be matched, the current translation result by the all translation means is translated into the target language. Output as result.

Further, in the automatic interpreting apparatus according to claim 2 of the present invention, the voice recognition means performs voice recognition of the uttered voice based on the input voice signal of the uttered voice of the source language. Is converted into a character string in the source language and output to the source sentence dividing means.

Further, in the automatic interpreting apparatus according to the third aspect of the present invention, the voice synthesizing means converts a character string of a translation result of a target language output from the translation generating means into a voice signal and outputs the voice signal. .

[0010]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 shows a block diagram of an automatic interpreter according to an embodiment of the present invention. The automatic interpreting apparatus of this embodiment translates a voice signal of a character string composed of an input utterance voice sentence of a source language into a character string of a target language, and then voice-synthesizes the character string resulting from the translation into a voice signal. 1. As shown in FIG. 1, a microphone 10 for converting (a) an input utterance voice of a source language into a voice signal;
(B) Speech recognition unit 1 that recognizes the uttered speech based on the input speech signal of the uttered speech in the source language, converts the uttered speech into a character string in the source language, and outputs the character string to the source sentence dividing unit.
And (c) a divided pattern storage unit 7 for storing a plurality of divided patterns in the uttered speech sentence of the source language in advance, and (d) a character string composed of the uttered speech sentence of the source language output from the speech recognition unit 1. An original sentence division unit 2 for dividing the character string by referring to a plurality of division patterns stored in the storage means;
(E) a partial translation unit 3 that translates a character string divided by the original sentence division unit 2 into a character string of a translation result of a target language for each divided unit; and (f) a utterance voice sentence in the source language. (G) the translation result of the partial translation unit 3 and the translation result of the translation result of the target language without dividing the A translation generator 5 that compares a character string and determines and outputs a character string of a translation result of a target language so as to omit redundant expressions.
(H) an output result storage unit 8 for storing a character string of the translation result of the target language output from the translation generation unit 5; and (i) a character string of the translation result of the target language output from the translation generation unit 5. Into a speech signal and output, and (j)
A speaker 11 is provided for converting an audio signal output from the audio synthesis unit 6 into audio and outputting the audio. The voice recognition unit 1
The apparatus excluding the earlier part and the part later than the speech synthesizer 6 is an automatic translation apparatus.

In this embodiment, an example is a translation from English to Japanese in a dialog having a clear purpose such as a "dialogue when making a hotel reservation", and a gradual process is performed when simultaneous interpretation is performed. Here is a simple sentence generation method. In other words, the input utterance is converted from English to Japanese in units such as phrases and clauses, using a delimiter expression such as a conjunction as a trigger, and the Japanese expression of the partial conversion result is sequentially determined while the entire sentence is determined. Here is a translation method to generate a natural Japanese sentence that makes sense.

FIG. 2 shows the processing time of the processing in units of a sentence and the stepwise processing in the automatic interpretation. In order to maintain the cohesiveness of the utterance and shorten the processing time, it is necessary to shorten the time until the output starts and the time until the output ends. As shown in FIG. 2, in the processing for each sentence, “time until output start = speech time + translation processing time”. Also, in many current translation apparatuses, the processing time is becoming faster, and “speech time≫translation processing time” is satisfied.
Therefore, shortening the translation processing time does not contribute much to shortening the time until output starts. On the other hand, if the gradual processing is performed, by performing the translation processing and the output before the utterance of the input ends, it is possible to greatly reduce the time until the output starts and the time until the output ends. .

For example, “It is 4000 yen,
if you registernext mont
h. Is translated, as shown in FIG.
By outputting the translation "It is 4000 yen" at the stage when "if" is uttered, it is possible to shorten the time until the start of utterance. In addition, the time required to output the final translation result "It's 4000 yen, if you register next month" can be shortened.

However, in a language such as Japanese and English having greatly different word order and structure, simply outputting the translation result in a forward order may generate a meaningless sentence. Further, when the final output is a voice, as in the case of an automatic interpreter, it is not possible to delete the output once, so a mechanism for self-repairing the output Japanese expression is required in the generation unit.

In order to progressively generate a Japanese dialogue sentence from an English sentence, the ability to generate a "grammatically qualified sentence", which has been the object of the conventional written word generation, is not sufficient. In written language, it is called "ineligible sentence", but in Japanese dialogue, "repetition"
There is "inversion" and the like. In the present embodiment, it is considered that the generation capability is enhanced by using these. Hereinafter, “repetition” and “inversion” will be briefly described.

In human speech, repetition may be used to add new information to an already mentioned phrase or correct an incorrect expression. For example, "Tell me a phone number and a phone number you can contact."
By repeating the "phone number", new information ("contact you") can be added to the phrase ("phone number") already mentioned.

In a Japanese sentence, it is said that the main verb of a clause (or sentence) must appear at the end of the sentence. However, in actual dialogues, words whose word order is inverted, such as the translation result in FIG. 3, are frequently seen. Here, in order to treat this phenomenon in the same framework as the above-mentioned repetition, it is considered that the word order is not caused by inversion, but is caused by repetition and omission of a sentence. For example, the sentence "It's 4000 yen, if you register next month"
"It's 4000 yen" is generated as a sentence with the condition omitted, then to clarify the condition of this sentence,
We consider that this sentence was generated by repeating the condition at the end of the sentence.

In order to incorporate the features of the above-mentioned Japanese dialogue sentence into the generation and avoid a decrease in the intelligibility of the generated sentence, in this embodiment, the translation is generated by the following method. (A) In order to ensure the intelligibility of the sentence, the word order of the dependency must always be from the source to the destination. (B) When the word order of the dependency is reversed, the sentence is self-repaired by repeating the part. (C) To reduce output time, redundant expressions in repetition are omitted.

FIG. 1 is a block diagram of an automatic interpreter using the above-described translation generating method. In FIG. 1, an uttered voice in a source language is input to a microphone 10, converted into a voice signal, and input to a voice recognition unit 1. The speech recognition unit 1 calculates the likelihood by executing phoneme matching based on the input speech signal of the uttered speech in the source language, for example, with reference to a phoneme context-dependent phoneme hidden Markov model (phoneme HMM), By selecting the maximum likelihood speech recognition result and performing speech recognition on the uttered speech, the speech is converted into a character string in the source language and output to the source sentence division unit 2. On the other hand, the division pattern storage unit 7 previously stores a plurality of division patterns in the utterance voice sentence in the source language as shown in Table 1, for example.

[0020]

[Table 1] English sentence patterns in English-Japanese translation and Japanese sentence patterns as translation results節 Clause type English pattern Japanese pattern ───────────────────────────── Sentence By the way ／ X By the way ／ X ' Sentence no / X No / X 'Parallel clause X / but Y X' / But Y 'Dependent clause X / if Y Y' // X 'Dependent clause if X / [CN-PRON] Y Y' / X 'Quotation section X think [V-PRON] / Y X' thinks Y 'Relative section X / where Y Y' [continuous] / X '───────────────── ────────────

Here, / indicates a division point. Also, [CN
[-PRON] means that the part of speech of the word before [CN-PRON] is a common noun and the part of speech of the word after [CN-PRON] is a pronoun, and [V-PRON] means [V-PRON].
[N] means that the part of speech of the word before is a verb, and the part of speech of the word after it is a pronoun. Further, [continuous] means that the inflected form of the preceding word is a continuous form.

In the present embodiment, the unit for dividing English sentences is (1) semantically united and easy to correspond between English and Japanese, and (2) relatively clear for dividing into English sentences. Because a marker (such as a conjunction) appears, it is a clause. As a pattern for dividing an English sentence into sections, a pattern used in an example-based translation device was used. Table 1 shows typical English sentence patterns. This English sentence pattern has a surface expression ("but",
"If" etc.) and part-of-speech sequence ([CN-PRON]
) Is used as a key to divide sentences into clauses. In Table 1, the number of the level indicates the strength of the coupling degree of the division, X and Y indicate the English expressions, and X ′ and Y ′ indicate the Japanese translations corresponding to the English expressions X and Y, respectively. .

The original sentence dividing unit 2 refers to the plurality of divided patterns stored in the storage unit and converts the character string composed of the uttered speech sentence of the original language output from the speech recognizing unit 1 into the character string. Is divided into appropriate units. Further, the partial translating unit 3 translates the character string divided by the original sentence dividing unit 2 into a character string of a translation result of the target language sequentially for each divided unit. Translation into a character string of the translation result of the target language is performed without dividing the character string composed of the spoken voice sentence of the language. Next, the translation generation unit 5
The character string of the translation result by the partial translation unit 3 is compared with the character string of the translation result by the full translation unit 4, and if necessary, it is repeated so as to omit redundant expressions, and in an appropriate word order. And determine and output the character string of the translation result of the target language. The output storage unit 8 stores the character string of the translation result of the target language output from the translation generation unit 5. The speech synthesis unit 6 is configured by a known speech synthesis device including, for example, a pulse generator, a noise generator, a variable amplification amplifier, and a filter, and a translation result of a target language output from the translation generation unit 5. Is converted to an audio signal and output to the speaker 11, and then converted to audio and output.

The partial translation unit 3 and the whole translation unit 4 can be configured using a known interactive sentence translation system using a bilingual example, for example. In this system, ("I wou
ldlike to ~ "→" I want to do it "),
Examples are prepared in relatively small units, such as (“for...” → “to...”), And by combining them, one sentence can be converted.

When the input sentence S is CL ₁ ,..., CL _{i ,.}
When the sentence is divided into (1 ≦ i ≦ j) sections, a sentence is translated and generated by the following method. However, CL _j, as shown in FIG. 4 is a right end sections of the currently input, CL _i is CL _j
Is the beginning clause of the largest clause containing. Also, let the translation result of the English CL be trans (CL).

The partial translation unit 3, the total translation unit 4, and the translation generation unit 5
Is repeated until the end of the utterance. The partial translation unit 3 obtains a partial translation result trans (CL) and outputs it to the translation generation unit 5, while the full translation unit 4 obtains a full translation result trans (CL _i ,..., CL _j ). 5 is output. Then, the translation generation unit 5 executes the translation generation processing shown in FIG. 6 as follows.

First, in step S1, the translation result t
trans (CL _j ) is the translation result trans (CL _i ,...,
CL _j ), and in the case of being equal, ie, i = j, in step S11, the translation result trans (CL _i ,..., CL _j ) is output as the translation result, and the translation generation processing is performed. finish. On the other hand, step S1
If i ≠ j, the translation result tr is obtained in step S2.
ans (CL _i ,..., CL _j−1 ) · trans (CL _j )
Is determined to be equal to the translation result (CL _i ,..., CL _j ). If so, in step S12, the translation result trans (CL _j ) is output as the translation result and the translation is performed. The generation processing ends. On the other hand, if the word order is different in step S2, in step S3, the translation result trans (CL _j ) · trans
(CL _i ,..., CL _j-1 ) is the translation result trans (C
L _i, ..., whether CL _j) equal or not. If equal, that is, when the word order is reversed, in step S13, the translation results trans (CL _i, ..., as a translation result CL _j), repeating the The common part (redundant expression) is omitted and output, and the translation generation process ends. On the other hand, if they are not equal in step S3, the translation result trans (C
L _i ,..., CL _j ) are output as translation results, and the translation generation processing ends.

Next, an example of the translation process will be described. "Cold I have your tele
phone number where I can
contact you? FIG. 5 shows the flow of the translation process, taking the sentence "as an example. First, "Cold I
have your telephone number
r where "is input, and" X where
The original sentence division unit 2 matches the pattern of the section “re Y”. Partial translation unit 3 and full translation unit 4
Performs partial translation and full translation (left part in FIG. 5),
In this example, the partial translation result and the full translation result are equal,
The translation generation unit 5 outputs “Tell me the telephone number” of all the translation results. Next, when the sentence is inputted to the end of the sentence, the partial translator 3 translates the part of “I can contact you”, and the all translator 4 translates the largest clause including the clause (that is, the whole sentence) (FIG. 5). Right part). in this case,
Since the word order is reversed, the translation generation unit 5 outputs "I can contact you" because the entire translation result is the translation result and the common part is omitted.

As described above, taking an English-Japanese translation as an example,
This paper shows a gradual sentence generation method when performing simultaneous interpretation. By using this method, it was shown that the time until the start of output of the translation result can be greatly reduced.
The automatic interpreting apparatus of the present embodiment according to the present invention can perform (1) a user's conversation without delay in transmission of an utterance by the interpreting process by performing the simultaneous interpreting translation process.
(2) There is an effect that the user can accomplish the purpose of the conversation in a short time.

In the above embodiments, the translation from English to Japanese has been described. However, the present invention is not limited to this, and the translation from Japanese to English and translation between two different languages are possible. Can be applied to

In the above embodiment, the speech recognition unit 1, the original sentence division unit 2, the partial translation unit 3, and the full translation unit 4
The translation generation unit 5 and the speech synthesis unit 6 are constituted by, for example, a digital computer.

In the above embodiment, a unit is used as a unit for dividing an English sentence as an original sentence. However, the present invention is not limited to this. Any unit and a combination thereof may be used.

[0033]

As described above in detail, according to the automatic translation apparatus of the first aspect of the present invention, an automatic translation for translating a character string composed of an input utterance voice sentence of a source language into a character string of a target language. In the translation device, a storage device that stores a plurality of division patterns in the utterance voice sentence of the source language in advance, and a character string composed of the input utterance voice sentence of the source language is converted into the plurality of division patterns stored in the storage device. An original sentence dividing unit that divides the character string, a partial translation unit that translates the character string divided by the original sentence dividing unit into a character string of a translation result of a target language for each divided unit, All the translation means for translating into the character string of the translation result of the target language without dividing the character string composed of the input utterance voice sentence of the source language, and the character string of the current translation result by the all translation means. A certain first string Comparing the character string of the previous translation result by the full translation means with a second character string obtained by concatenating the character string of the current translation result by the partial translation means, If the character string matches the second character string and the word order is the same, the character string of the current translation result by the partial translation means is output as the translation result of the target language. After the character string of the current translation result by the partial translation means,
A third character string obtained by concatenating the character strings of the immediately preceding translation result is compared. If the first character string matches the third character string and the word order is reversed, The translation result obtained by omitting redundant expressions from the current translation results by the translation means is output as the translation result of the target language, while the first character string does not match the third character string, and the word order is not matched. A translation generating unit that outputs a current translation result of the all translation unit as a translation result of a target language when the translation result cannot be obtained. Therefore, the translation can be performed in a shorter time than in the conventional example, and furthermore, the sentence can be translated into a natural sentence so that the entire sentence makes sense.

[0034] According to the automatic translating apparatus according to the second aspect of the present invention, in the automatic translating apparatus according to the first aspect, the utterance speech is input based on the input speech signal of the source language utterance. The apparatus further includes a voice recognition unit that converts the data into a character string in the source language by voice recognition and outputs the converted character string to the source sentence division unit. Therefore, according to this automatic interpreter,
It is possible to translate the uttered voice uttered by the other party and perform a translation process of simultaneous interpretation, whereby (1) the user can perform a conversation without delay in transmission of the utterance by the translation process, (2) There is a unique effect that the user can accomplish the purpose of the conversation in a short time.

According to a third aspect of the present invention, there is provided the automatic interpreting apparatus according to the second aspect, wherein the character string of the translation result of the target language output from the translation generating means is converted into an audio signal. And a voice synthesizing means for converting the output into a speech. Therefore, according to this automatic interpreter,
It is possible to translate the uttered voice uttered by the other party, synthesize the voice, and perform a simultaneous interpreting translation process.
There is a specific effect that the user can perform a dialogue without delay in transmission of the utterance by the interpreting process, and (2) the user can achieve the purpose of the dialogue in a short time.

[Brief description of the drawings]

FIG. 1 is a block diagram of an automatic interpreting apparatus according to an embodiment of the present invention.

FIG. 2 is a diagram illustrating a difference in processing time between a sentence unit processing method and a progressive processing method.

FIG. 3 is a diagram showing an example of translation using a progressive processing method.

FIG. 4 is a diagram illustrating processing of a clause in a translation generation unit in FIG. 1;

FIG. 5 is a diagram showing a translation process of a relative clause in the automatic interpreting apparatus of FIG. 1;

FIG. 6 is a flowchart illustrating a translation generation process executed by the translation generation unit of FIG. 1;

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Voice recognition part, 2 ... Original sentence division part, 3 ... Partial translation part, 4 ... Full translation part, 5 ... Translation generation part, 6 ... Speech synthesis part, 7 ... Division pattern storage part, 8 ... Output result storage part, 10: microphone, 11: speaker.

────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-5-61905 (JP, A) JP-A-5-151256 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) G06F 17/28 G10L 3/00

Claims (3)

(57) [Claims] 1. An automatic translator for translating a character string composed of an input utterance voice sentence in a source language into a character string in a target language, comprising: a storage device for storing in advance a plurality of divided patterns in the source language utterance voice sentence; A character string composed of an input utterance voice sentence in a source language is referred to by referring to a plurality of divided patterns stored in the storage device,
An original sentence dividing unit for dividing the character string; a partial translation unit for translating the character string divided by the original sentence dividing unit into a character string of a translation result of a target language for each divided unit; All translation means for translating a character string composed of the utterance voice sentence of the source language into a character string of a translation result of the target language without dividing the character string; Comparing the character string with a second character string obtained by concatenating the character string of the current translation result by the partial translation means after the character string of the previous translation result by the all translation means,
When the first character string matches the second character string and the word order is the same, the character string of the current translation result by the partial translation means is output as the translation result of the target language,
The first character string is compared with a third character string obtained by concatenating the character string of the current translation result by the partial translation means with the character string of the previous translation result by the all translation means. If the first character string and the third character string match and the word order is reversed, the translation result obtained by omitting redundant expressions from the current translation results by the all translation means is converted to the target language. While outputting as a translation result, if the first character string does not match the third character string and the word order cannot be matched, the current translation result by the all translation means is output as the translation result of the target language. An automatic translation device, comprising: 2. The automatic translation apparatus according to claim 1, wherein said uttered voice is speech-recognized on the basis of an input speech signal of the uttered voice in the source language to convert it into a character string in the source language. An automatic interpreting apparatus, further comprising a voice recognition means for outputting to a dividing means. 3. The automatic interpreting apparatus according to claim 2, further comprising a voice synthesizing means for converting a character string of a translation result of the target language output from said translation generating means into a voice signal and outputting the voice signal. Automatic interpreter.