JPS6321948B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to improvements in electronic translation devices.

Note that in this specification, translation refers to (1) translating the language of one country into the language of another country, (2) mutual conversion of an abbreviation and the original text even if they are in the same country, for example, Mutual conversion between standard language and dialect, mutual conversion between letters and words (sounds), conversion between words and their explanations, etc. (3) Arithmetic operations (general calculations and adding “π” to “3.14… ”).

The source language refers to the language, numbers, formulas, etc. before translation, and the translated language refers to the language, numbers, formulas, etc. after translation. For example, when an electronic translation device is set up to translate Japanese to English,
is the original language, and "gutsudmorning" is the translated language.

FIG. 1 is a block diagram of an electronic translation device already proposed by the present inventor in Japanese Patent Application Laid-Open No. 56-103765.

In the figure, reference numeral 1 denotes an input terminal for applying an original language input signal, which is connected to an electronic translation circuit 4 set to Japanese-English via an input amplifier 2 and a speech analysis circuit 3. Reference numeral 7 denotes a speaker connected to the electronic translation circuit 4 via the speech synthesis circuit 5 and the output amplifier 6 for transmitting speech output of the translated language. 8 is
This is a character display unit connected to the electronic translation circuit 4 and made of liquid crystal or the like for displaying characters in the original language and the translated language. S1 is a translation instruction switch for causing the electronic translation circuit 4 to translate the original language applied from the input terminal 1.

Next, the operation of this device will be explained.

A speech input signal in the original language applied to the input terminal 1 is amplified by the input amplifier 2 to a level that allows speech analysis, and then sent to the speech analysis circuit 3. The speech analysis circuit 3 performs spectrum analysis on the applied speech signal, performs speech recognition, converts it into a character signal of the original language, and stores the character signal in the memory of the electronic translation circuit 4. Next, by manually turning on the translation instruction switch S1 and applying a translation instruction pulse to the electronic translation circuit 4, the electronic translation circuit 4 reads out the character signal of the source language previously stored in its own memory, and is translated into English, and a character signal in the translated language is sent to the speech synthesis circuit 5. The speech synthesis circuit 5 is
The applied character signal is converted into a voice signal of the translated language (1) and sent to an output amplifier (6). Output amplifier 6
amplifies this audio signal and, via the speaker 7, produces audio in the translated language corresponding to the original language audio signal applied from the input terminal 1 earlier. Further, the character display section 8 displays characters of the source language and characters of the translated language applied to the electronic translation circuit 4.

In this manner, in the electronic translation device described above, only the audio output of the translated language is amplified from the speaker 7. For this reason, when the input terminal of such an electronic translation device is connected to the audio output terminal of a television, radio telephone, etc., the original language can be heard in the translated state from the speaker, but the surroundings such as background music can be heard. Sound could not be heard through the speakers.

The present invention eliminates such drawbacks,
The present invention provides an electronic translation device capable of extracting an input signal applied to an input terminal of a source language as well as a translated language output.

An embodiment of the present invention will be described below with reference to FIG.

FIG. 2 is a block diagram of an electronic translation device which is an embodiment of the present invention.

1 in the figure is an input terminal for inputting the original language. 2
is an input amplifier connected via a speech analysis circuit 3 to an electronic translation circuit 4 set to English-Japanese.
Input amplifier 2 is connected to input terminal 1.

Here, the speech analysis circuit is a circuit that spectrally analyzes a speech input signal, performs speech recognition, and sends out a character signal according to the speech input signal.

The electronic translation circuit 4 is connected to one fixed terminal a of a variable resistor (hereinafter referred to as a volume) 11 via a speech synthesis circuit 5.

Here, the speech synthesis circuit is a circuit that converts a character signal into a speech signal, and the speech synthesis method includes:
formant synthesis method, linear predictive coding (LPC)
Three types of methods have been put into practical use: waveform, waveform, and digitization/compression methods. In this embodiment, the Parcor method, which has particularly stable synthesis characteristics among linear predictive coding (LPC) methods, is used in the speech synthesis circuit 5. The other fixed terminal C of the volume 11 is connected to a delay circuit 10 whose delay time is variable (including zero delay time).
It is connected to the output terminal of the input amplifier 2 via. A sliding terminal b of the volume 11 is connected to a speaker 7 via an output amplifier 6. 9 is
This is an audio operation control circuit (hereinafter referred to as a VOX circuit) connected to the output terminal of the input amplifier 2. VOX
As already proposed by the present applicant in Japanese Patent Application No. 176296/1980, the circuit 9 filters the signal applied to the input terminal 1 through a band pass filter with a passband frequency of 70 to 2700 Hz, and filters the audio signal of the original language. is activated when a no-signal state continues for a certain period of time (approximately 1.0 seconds in this embodiment) in the audio signal, short-circuits the translation instruction switch S2, and sends the translation instruction pulse to the electronic translation circuit 4.
and causes the original language applied before the no-signal state to be translated.

Further, as described above, the translation instruction switch S2 is configured to be opened and closed not only automatically but also manually.

Next, the operation of this device will be explained.

Now, if input terminal 1 is connected to a radio receiver (not shown) that is "receiving a radio broadcast of a man walking down the street and asking directions to people passing by" think of. Input terminal 1 from the radio receiver is the man's
A sound signal of ``city noise such as cars'' is applied along with a voice signal of ``Me!''.

The voice signal and sound signal applied to the input terminal 1 are both amplified by the input amplifier 2 and applied to the voice analysis circuit 3.The voice analysis circuit 3 converts the voice signal "Excuse me" into "EXCUSE".
It only converts to a character signal "ME.", so
No sound signal is sent from the speech analysis circuit 3 to the electronic translation circuit 4.

The VOX circuit 9 is activated when silence continues for about 1 second after the voice signal (provided that the level of the sound signal is sufficiently low compared to the level of the voice signal).
The translation instruction switch S2 is turned on, and a translation instruction pulse is sent to the electronic translation circuit 4 to cause the electronic translation circuit 4 to perform a translation operation. The electronic translation circuit 4 translates the character signal "EXCUSE ME." in the source language into the character signal "SUMIMASEN" in the translated language, and sends it to the speech synthesis circuit 5.

The speech synthesis circuit 5 converts this character signal into a speech signal "excuse me" and applies it to the fixed terminal a of the volume 11.

On the other hand, the voice signal "excuse me" sent from the input amplifier 2 and the sound signal such as car noise are delayed for a certain period of time by the delay circuit 10 and then applied to the fixed terminal C of the volume 11. Therefore, as the sliding terminal b is brought closer to the fixed terminal a, the output signal appearing on the sliding terminal b of the volume control 11 is such that as the sliding terminal b is brought closer to the fixed terminal a, the voice signal of the translated language, "I'm sorry," becomes more pronounced than the voice signal of the original language, "I'm sorry.""ExcuseMe" and the sound signal, and as the sliding terminal B approaches the fixed terminal C, the voice signal of the source language and the sound signal become louder than the voice signal of the translated language, contrary to the above. Can be done. That is, the amount of mixing between the output of the translated language and the output of the original language can be changed depending on the position of the sliding terminal b. The output signal mixed in this manner is amplified by the output amplifier 6 and then amplified as audio from the speaker 7.

At this time, by appropriately adjusting the volume 11 from the speaker 7, it is possible to amplify not only the voice of the translated language but also the voice of the original language and surrounding sounds.

Furthermore, the delay time of the delay circuit 10 can be determined by voice analysis.
The time is set to be approximately equal to the sum of the times required for electronic translation and speech synthesis (approximately 1 to 3 seconds). By doing so, it is possible to substantially match the transmission times of the output in the original language and the output in the translated language.

When such a delay is not necessary, for example, when practicing translation using a language tape, you can listen to the source language, translate it yourself, then hear the translated language over a loudspeaker, and as a result judge whether your translation is good or bad. Since the procedure is good, in such a case, it is better to adjust the delay circuit and make the delay time zero.

Next, another embodiment of the present invention will be described using FIG.

FIG. 3 is a block diagram of an electronic translation device according to another embodiment of the present invention.

In the figure, reference numeral 1 denotes an input terminal for inputting the original language, which is connected to the electronic translation circuit 4 via the input amplifier 2 and the speech analysis circuit 3. The electronic translation circuit 4 is connected to a speaker 7 via a speech synthesis circuit 5 and an output amplifier 6. Reference numeral 9 denotes a VOX circuit for driving the translation instruction switch S2, which is the same as in the previous embodiment. 10 is a delay circuit connected to the input amplifier 2, and the delay circuit 10 is connected to the input terminal of the output amplifier 6 via a voltage control amplifier (hereinafter referred to as a VCA circuit). 12 is a VOX circuit connected to the electronic translation circuit 4 and the VCA circuit 13 for controlling the amplification degree of the VCA circuit 13.

The electronic translation device of this embodiment is the same as that of the previous embodiment except for the VOX circuit 12 and the VCA circuit 13, and operates in the same manner as described above. Therefore, the relationship between the VOX circuit 12 and the VCA circuit 13 will be explained in detail.

The VOX circuit 12 is connected to the VCA circuit 12 while the electronic translation circuit 4 is sending out character signals in the translated language and for a certain period of time (approximately 1 second) after the sending is completed.
control so that the degree of amplification is reduced. This results in
While the speech signal in the translated language is being sent from the speech synthesis circuit 5 to the output amplifier 6, the level of the speech signal in the original language applied from the VCA circuit 13 to the output amplifier 6 decreases. Therefore, while the translated language audio circuit is being transmitted from the speaker 7, the output level of the original language is lowered and the translated language becomes easier to hear. In the above, the reason why the control period of the VCA circuit 13 is set to approximately 1 second after the completion of sending the character signal of the translated language is that approximately 1 second is required for speech synthesis.

In the above two embodiments, an example was shown in which the output of the original language and the output of the translated language were amplified using the same output amplifier 6 and speaker 7, but the present invention is not limited to this. The translated language output may be amplified using a separate and independent output circuit (consisting of an output amplifier and a speaker).

Further, an input amplifier connected to the input terminal 1 may be provided separately from the input amplifier 2 for outputting the original language.

Note that when using such a device in a television receiver, a delay circuit is also inserted in the image signal output circuit.
It is a good idea to delay both audio and images.

In the embodiment of the present invention, an example was shown in which the VOX circuit 9 was connected to the output terminal of the input amplifier 2, but this is connected to the output terminal of the speech analysis circuit 3, and the VOX circuit 9 is activated only for the voice signal of the original language. The VOX circuit 9 can also be activated only when no voice signal in the original language is applied for a certain period of time or more. Also, this is the VOX circuit 9
can be activated exactly at the end of the voice signal without being interfered with by other ambient sounds or other sound signals. Further, by stopping the VOX circuit 9 and keeping the translation instruction switch S2 open, only the audio signal applied to the input terminal 1 can be heard from the speaker 7 without being translated. By connecting a telephone pickup to the input terminal 1, electromagnetic coupling can be established with the telephone receiver, so that the voice signal of the other party can be accurately translated without being interfered with by surrounding sounds. Furthermore, if a microphone is connected to the input terminal 1, it can also be used as a hearing aid. Note that the VOX circuit, delay circuit, and VCA circuit used in the embodiments of the present invention may be configured on a single-chip LSI together with a speech recognition circuit, an electronic translation circuit, a speech synthesis circuit, etc. A microcomputer used in a circuit or a speech synthesis circuit can also be used to control and process a series of operations such as translation instructions, level settings for the VCA circuit, and control time settings for the circuit.

As described above, according to the present invention, by providing a branching means for branching the source language information and an addition means for adding the source language information before translation and the translated language information after translation branched by the branching means, , the original language information and the translated language information can be extracted in a mixed state, making it possible to perform more natural translation and to compare both language information.

Further, if the present invention is applied to an audio input/output type electronic translation device, it is possible to transmit to the audio information output side, bird songs, music, etc. that cannot be translated.

Therefore, if such electronic translation devices are used in television receivers, radio receivers, telephone equipment, etc., people on the receiving end will not only hear the voice of the translated language, but will also be able to hear the surrounding sounds of the sending end, background music, etc. Since it is also possible to convey the voice of the source language (the voice of the person on the sender's side), it is possible to convey more natural translations, the emotional expression of the speaker, and the circumstances around the speaker.

[Brief explanation of drawings]

FIG. 1 is a block diagram of a conventional electronic translation device, FIG. 2 is a block diagram of an electronic translation device that is an embodiment of the present invention, and FIG. 3 is a block diagram of an electronic translation device that is another embodiment of the present invention. It is a diagram. 1...Input terminal, 2...Input amplifier, 3...Speech analysis circuit, 4...Electronic translation circuit, 5...Speech synthesis circuit, 6...Output amplifier, 7...Speaker, 8
... Character display section, 9, 12 ... Audio operation control circuit (VOX circuit), 10 ... Delay circuit, 11 ... Variable resistor for mixing (volume), 13 ... Voltage control amplifier (VCA circuit), S1 , S2... Translation instruction switch.

Claims (1)

[Scope of Claims] 1. Information input means for inputting information regarding the source language to be translated; branching means for branching the information input into the information input means into at least two parts; and a source branched by the branching means. electronic translation means for translating into information about the corresponding translation language based on the information about the language; and adding the information about the translation language translated by the electronic translation means and the information about the original language branched by the branching means. An electronic translation device comprising: an addition means; and an information output means for outputting both pieces of information added by the addition means. 2. The electronic translation means includes a voice recognition means that recognizes voice information regarding the source language, and an electronic translation unit that translates the information regarding the source language recognized by the voice recognition means into information regarding the corresponding translation language, based on the information regarding the source language recognized by the voice recognition means. 2. The electronic translation device according to claim 1, further comprising speech synthesis means for synthesizing speech information corresponding thereto based on information regarding the translated language translated by the electronic translation section. 3. The electronic translation device according to claim 1, wherein the addition means includes variable means for changing the addition ratio of both information signals. 4. The electronic translation device according to claim 3, wherein the addition means operates variable means for changing the addition ratio of both information signals while the translation output is being sent from the electronic translation means. 5. Claims 1 to 4, characterized in that the adding means has a control means for controlling the output timing of the output information from the branching means and the output information from the electronic translation means to substantially match each other.
The electronic translation device according to any of paragraphs.