JP2010026220A - Voice translation device and voice translation method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide voice translation device that facilitates performing voice translation of bidirectional interaction by one device. <P>SOLUTION: The voice translation device includes: a first microphone for collecting voice by a first language; a first translation section for performing translation in which voice expressed by a signal output by the first microphone is converted to a second language; a first speaker section for output of voice of the signal which is output by the first translation section; a second microphone for collecting voice by the second language; a second translation section for performing translation in which voice expressed by a signal output by the second microphone is converted to the first language; and a second speaker for output of voice of the signal which is output by the second translation section. A direction of a sound source, in which the first microphone most effectively collects voice is about the same as a direction in which the second speaker outputs voice, and a direction of a sound source, in which the second microphone most effectively collects voice is about the same as a direction in which the first speaker outputs voice. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a speech translation apparatus and a speech translation method.

  A conventional speech translation apparatus includes a sound collection microphone in addition to a translation input microphone, and determines the surrounding environment based on the sound input from the sound collection microphone, and is input from the translation input microphone using the determined ambient environment. The translated voice is translated and the translated voice is output from the translation output speaker (for example, Patent Document 1).

In Patent Document 2, an audio signal in an audible band is modulated into a signal in a frequency band (ultrasonic band) higher than the audible band, and output from a speaker, so that an audio signal having a directivity is output. Along with the directional speaker technology, this super directional speaker technology allows a television receiver displaying a plurality of programs to output the sound of each program in different directions. Is described. There is also a double-sided video display device provided with a super-directional speaker corresponding to each surface of the double-sided display (for example, Patent Document 3).
In addition, there is a mobile terminal device such as a mobile phone terminal that includes a plurality of super-directional speakers so that the contents of a call cannot be heard by anyone other than the user (for example, Patent Document 4).
JP 2003-108551 A Japanese Patent Laid-Open No. 11-262084 JP 2007-251331 A JP 2006-067386 A

However, in the conventional speech translation apparatus, in order to perform a two-way dialogue, it is necessary to use the speech translation apparatus for the number of the talkers, or one speech translation apparatus between the talkers. There is a problem that it must be used alternately and is complicated.
In addition, although the television receiver of patent document 2 and the double-sided display of patent document 3 output audio | voice individually to several persons, a dialog and voice translation cannot be performed, and the mobile telephone terminal of patent document 4 is voice-translated. Although it is not possible to perform two-way dialogue, it is necessary to have a device equivalent to the number of dialoguers.

  The present invention has been made in view of such circumstances, and an object of the present invention is to provide a speech translation apparatus capable of easily performing speech translation of a two-way dialogue with a single device.

(1) The present invention has been made to solve the above-described problems, and the speech translation apparatus of the present invention collects speech in the first language, converts the collected speech into a signal, and outputs the signal. A first microphone that receives the signal output from the first microphone, performs a translation process to generate a signal representing a voice converted from the voice represented by the signal into a second language, and outputs a signal of the translation process result A first translation unit, a first speaker unit that outputs a voice of a signal output from the first translation unit, a voice in a second language is collected, and the collected voice is converted into a signal and output. Receiving a signal output from the second microphone, performing a translation process for generating a signal representing a voice converted from the voice represented by the signal into the first language, and outputting a signal of the translation process result A second translation unit, and the second translation unit A second speaker unit that outputs the sound of the signal output by the translation unit, and the direction of the sound source that the first microphone can collect the sound most efficiently and the direction in which the second speaker unit outputs the sound. The direction of the sound source that can be picked up most efficiently by the second microphone and the direction in which the first speaker unit outputs sound are substantially the same.

(2) The speech translation device of the present invention is the speech translation device described above, wherein the first speaker unit and the second speaker unit are directional speakers that output speech with directivity. It is characterized by that.

(3) Further, the speech translation device of the present invention is the speech translation device described above, wherein the second speaker unit is receiving a signal to be translated from the second microphone. A first translation state notifying unit for notifying in a direction of outputting sound, and that the first translation unit is receiving a signal to be translated from the first microphone; And a second translation state notifying unit for notifying.

(4) Moreover, the speech translation apparatus according to the present invention is the above-described speech translation apparatus, and notifies the second speaker unit in the direction of outputting speech that the first translation unit is performing translation processing. A first translation state notifying unit and a second translation state notifying unit for notifying in a direction in which the first speaker unit outputs a voice that the second translation unit is performing a translation process, To do.

(5) Further, the speech translation device of the present invention is the speech translation device described above, wherein the first speaker unit is outputting a speech signal as a translation processing result, and the second speaker unit The first translation state notifying unit for notifying in the direction of outputting the signal and the second translation unit notifying that the first speaker unit outputs the sound signal of the translation processing result in the direction of outputting the sound. And a second translation state notifying unit.

(6) Moreover, the speech translation device of the present invention is the speech translation device described above, wherein the first speaker unit includes the second translation unit in addition to the speech of the signal output from the first translation unit. The sound of the output signal is output, and the second speaker unit outputs the sound of the signal output from the first translation unit in addition to the sound of the signal output from the second translation unit. .

(7) The speech translation apparatus of the present invention is the speech translation apparatus described above, wherein the first translation unit further receives a signal resulting from translation processing by the second translation unit, and the speech represented by the signal Is translated into a second language to generate a signal representing the speech, and a signal resulting from the translation process is output. The second translation unit is a signal resulting from the translation process performed by the first translation unit. And performing a translation process for generating a signal representing a voice obtained by converting the voice represented by the signal into a first language, and outputting a speech signal as a result of the translation process.

(8) Moreover, the speech translation apparatus of the present invention is the speech translation apparatus described above, wherein the first translation start instruction unit receives a translation processing start instruction by the first translation unit, and the translation by the second translation unit. A second translation start instruction unit that receives a process start instruction, and the first translation unit receives a signal output from the first microphone when the first translation start instruction unit receives an instruction to start translation. The second translation unit receives a signal output from the first microphone when the second translation start instruction unit receives an instruction to start translation, and receives a signal output from the first microphone. The translation processing for converting to the language is started.

(9) The speech translation device of the present invention is the speech translation device described above, wherein the second translation start instruction unit includes an imaging unit that images the output direction of the first speaker unit, and the imaging unit The translation start instruction is detected from the imaging result obtained by.

(10) The speech translation device of the present invention is the speech translation device described above, wherein the second translation start instruction unit receives a signal output from the second microphone and instructs the translation start from the signal. Is detected.

(11) In the speech translation method of the present invention, in the speech translation method in the speech translation device, the first microphone of the speech translation device collects the speech in the first language, and the collected speech is A first step of converting the signal into a signal and outputting the signal, and a signal representing the sound obtained by converting the speech represented by the speech translation device into the second language upon receiving the signal output in the first step. A second step of performing a translation process to be generated and outputting a signal of the result of the translation process; and a voice of the signal output by the speech translation device in the second step is a second microphone of the speech translation device. Is output in a direction substantially coincident with the direction of the sound source that can collect sound most efficiently, and the second microphone of the speech translation apparatus collects speech in the second language, The converted voice is converted into a signal and output. And a translation process in which the speech translation device receives the signal output in the fourth step and generates a signal representing speech obtained by converting speech represented by the signal into a first language, The fifth process of outputting the processing result signal, and the voice of the signal output by the speech translation apparatus in the fifth process are approximately the direction of the sound source that the first microphone can collect most efficiently. And a sixth process of outputting in the matching direction.

  According to this invention, the direction of the sound source that the first microphone can collect sound most efficiently coincides with the direction in which the second speaker unit outputs sound, and the sound source that the second microphone can collect sound most efficiently. Since the direction in which the first speaker unit outputs the voice substantially coincides with each other, the voice translation of the interactive dialogue can be easily performed with one unit.

[First Embodiment]
Hereinafter, a first embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a schematic block diagram showing the configuration of a speech translation apparatus 1 according to the first embodiment of the present invention. The speech translation apparatus 1 includes a first microphone 10, a first translation unit 11, a first speech storage unit 12, a first directional speaker unit 13, a first translation state notification unit 16, a display unit 17, and a first translation start instruction unit. 18, 2nd microphone 21, 2nd translation part 22, 2nd audio | voice storage part 23, 2nd directional speaker part 24, 2nd translation state notification part 27, LED (Light Emitting Diode) 28, 2nd translation A start instruction unit 29 is provided. The speech translation apparatus 1 can perform speech translation between various languages depending on the setting. In the following, a conversation person A who understands the first language and a conversation person B who understands the second language. A case where a two-way dialogue with a voice is translated is explained.

  The first translation start instructing unit 18 receives from the outside an instruction to start translation processing by the first translating unit 11, that is, speech translation processing of the conversation person A. In the present embodiment, the first translation start instruction unit 18 is provided on the surface of the speech translation apparatus 1 and is an operation unit (detection) that is a key button (a specific area on the outer surface) that detects being pressed (pressurization). Part) 20 and a first instruction detection part 19 for detecting that a key button of the operation part 20 is pressed as a translation process start instruction. Here, there are a plurality of key buttons, and each of the numbers from 0 to 9 is associated with “#”, “*”, “decision”, and the like. , It may be detected that one of these has been pressed as an instruction to start translation processing, or that a specific one, for example, a “decision” key button has been pressed It may be detected as a start instruction. In the present embodiment, the operation unit 20 is described as a key button that is provided on the surface of the speech translation apparatus 1 and detects that it is pressed. However, the operation unit 20 is in contact with a specific area on the outer surface as in a touch panel. May be detected.

The first microphone 10 is a microphone that includes a human voice frequency band in a frequency band where sound can be collected. The first microphone 10 collects a voice in the first language from the outside of the conversation person A, and signals the collected voice as a signal. Convert to and output.
The first translation unit 11 receives a signal output from the first microphone 10, that is, a voice signal of the conversation person A, and performs a translation process for generating a signal representing the voice obtained by converting the voice represented by the signal into the second language. The translation processing result signal is output. More specifically, when the first translation start instructing unit 18 receives an instruction to start translation, the first translation unit 11 receives a signal output from the first microphone 10, performs speech recognition processing, and represents the voice represented by the signal. Is converted into a character string of the first language and stored in the first speech storage unit 12. Next, the 1st translation part 11 converts the character string of the 1st language memorize | stored in the 1st audio | voice storage part 12 into the character string of a 2nd language by performing a translation process. Further, the first translation unit 11 performs a speech synthesis process to generate and output a second language speech signal from the second language character string. The first sound storage unit 12 is a readable / writable memory such as a RAM (Random Access Memory) or a hard disk.

  The first directional speaker unit (first speaker unit) 13 outputs the voice of the signal output from the first translation unit 11, that is, the voice obtained by translating the voice of the first language by the conversation person A into the second language. . In the present embodiment, the first directional speaker unit 13 includes a first directional control unit 14 and a first speaker 15 and functions to output sound with directivity. The first directivity control unit 14 modulates the voice signal (audio signal in the audible band) received from the first translation unit 11 into a signal in a frequency band (ultrasonic band) higher than the audible band. The first speaker 15 is an ultrasonic wave generating element that can output ultrasonic waves in the frequency band of the modulation result by the first directivity control unit 14. The first speaker 15 outputs the ultrasonic wave of the signal modulated by the first directivity control unit 14 to output the sound modulated by the first directivity control unit 14 with directivity. In this way, by modulating the audio signal into a signal in a frequency band (ultrasonic band) higher than the audible band and outputting the ultrasonic wave of the signal with a speaker, it is possible to output sound with directivity. There are known techniques, and the modulation method used at this time includes amplitude modulation, frequency modulation, phase modulation, and the like, and any of them may be used.

  The second translation start instructing unit 29 receives an instruction to start translation processing by the second translating unit 22, that is, a speech translation processing of the conversation person B from the outside. In the present embodiment, the second translation start instructing unit 29 is provided on the surface of the speech translation apparatus 1, and is a camera (imaging unit) that captures an image of the voice output direction of the first speaker (that is, images the conversation person B). ) 31 and the second instruction detection unit 30. The second instruction detection unit 30 detects detection of a predetermined shape or subject movement from an image captured by the camera 31 as a start instruction. Thereby, when the conversation person A carries the speech translation apparatus 1, the conversation person B starts translation by taking a specific pose (predetermined shape) or performing a specific movement. Can be directed.

The second microphone 21 is a microphone that includes a human voice frequency band in a frequency band in which sound can be collected. The second microphone 21 collects a voice in a second language from the outside by the conversation person B, and signals the collected voice as a signal. Convert to and output.
The second translation unit 22 receives a signal output from the second microphone 21, that is, a voice signal of the conversation person B, and performs a translation process for generating a signal representing the voice obtained by converting the voice represented by the signal into the first language. The translation processing result signal is output. More specifically, when the second translation start instruction unit 29 receives an instruction to start translation, the second translation unit 22 receives a signal output from the second microphone 21, performs voice recognition processing, and represents the voice represented by the signal. Is converted into a character string of the first language and stored in the second speech storage unit 23. Next, the second translation unit 22 performs a translation process to convert the second language character string stored in the second voice storage unit 23 into a first language character string. Further, the second translation unit 22 performs speech synthesis processing to generate and output a first language speech signal from the first language character string. The second sound storage unit 23 is a readable / writable memory such as a RAM (Random Access Memory) or a hard disk.

  The second directional speaker unit (second speaker unit) 24 outputs the voice of the signal output from the second translation unit 22, that is, the voice obtained by translating the voice of the second language by the conversation person B into the first language. . In the present embodiment, the second directivity speaker unit 24 includes a second directivity control unit 25 and a second speaker 26, and serves to output sound with directivity. Similar to the first directivity control unit 14, the second directivity control unit 25 converts the audio signal (audio signal in the audible band) received from the second translation unit 22 into a frequency band (ultrasonic band) higher than the audible band. ) Signal. The second speaker 26 is an ultrasonic wave generating element similar to the first speaker 15 that can output ultrasonic waves in the frequency band of the modulation result by the second directivity control unit 25. The second speaker 26 outputs the ultrasonic wave of the signal modulated by the second directivity control unit 25, thereby outputting the sound modulated by the second directivity control unit 25 with directivity. As described in the description of the first directional speaker unit 13, it is a known technique to output sound with directivity in this way. The modulation method used at this time includes amplitude modulation, There are frequency modulation, phase modulation, and the like, and any of them may be used.

  The first translation state notifying unit 16 is notified from the first translation unit 11 and the second translation unit 22 to acquire their operation states, and the second translation unit 22 receives the second microphone 21 in accordance with the acquired states. That the signal to be translated is being received from the first translation unit 11, that the first translation unit 11 is in translation processing, and that the first translation unit 11 is outputting the speech signal of the translation processing result, that is, second That the speaker 26 is outputting the voice of the translation result is displayed on the display unit 17 which is a display such as a liquid crystal or organic EL (Organic Electro-Luminescence). And notify.

  The display unit 17 is a display that displays in a direction in which the second directional speaker unit 24 outputs sound. That is, the display surface of the display unit 17 is provided on the surface of the speech translation apparatus 1 so as to face the same direction as the second speaker 26 of the second directional speaker unit 24, and the second directional speaker unit 24 outputs it. The display content of the display unit 17 can be viewed from the position where the received voice can be heard.

  The second translation state notification unit 27 is notified from the first translation unit 11 and the second translation unit 22 to acquire the operation state thereof, and the first translation unit 11 performs the first microphone 10 according to the acquired state. That the signal to be translated is being received from the second translation unit 22, that the second translation unit 22 is in the translation process, and that the second translation unit 22 is outputting the speech signal of the translation process result, that is, the first The speaker 15 is notified by causing the LED 28 to emit light with a light emission pattern such as a color indicating that the speech of the translation result is being output or blinking, but the mode of this notification is not limited thereto.

The LED 28 displays in the direction in which the first directional speaker unit 13 outputs sound, that is, the speech translation so that the display content can be seen from the position where the sound output from the first directional speaker unit 13 can be heard. It is provided on the surface of the device 1.
In addition, the direction of the sound source that the first microphone 10 can collect sound most efficiently coincides with the direction in which the second speaker 26 of the second directional speaker unit 24 outputs sound, and the second microphone 21 is the most efficient. The direction of the sound source that can collect sound well and the direction in which the first speaker 15 of the first directional speaker unit 13 outputs sound substantially coincide.

  Since the first microphone 10, the second speaker 26, the second microphone 21, and the first speaker 15 are arranged in this way, the conversation person A is translated into the first language output from the second speaker 26. What is necessary is just to speak in a 1st language toward the 1st microphone 10 in the position which can hear an audio | voice. The conversation person B may speak in the second language toward the second microphone 21 at the position where the voice translated into the second language output from the first speaker 15 can be heard. For this reason, speech translation of interactive dialogue can be easily performed with one unit.

  In addition, since the display unit 17 displays the second speaker 26 of the second directional speaker unit 24 in a direction in which sound is output, the conversation person A is translated into the first language output from the second speaker 26. If the user is at a position where he can hear the voice, the notification by the first translation state notification unit 16 displayed on the display unit 17 can be seen to recognize the state of the speech translation. That is, among the notifications by the first translation state notification unit 16, the second translation unit 22 recognizes the notification that the signal to be translated is being received from the second microphone 21. It can be recognized that the content spoken by the conversation partner B is translated and output from the second speaker 26 later.

  In addition, by recognizing that the first translation unit 11 is in the process of translation, the conversation person A is translating the contents that he has spoken earlier and is still being communicated to the conversation person B, the conversation partner. I can recognize that there is no. Further, by recognizing that the second speaker 26 is outputting the voice of the translation result, the conversation person A can hear the content that the conversation partner B talks about first. Recognize that it is inside. Thus, as in this embodiment, even if the translation result cannot be heard because a directional speaker is used, the speaker can be notified of the translation state of the speech translation apparatus 1 and whether or not the translation result is notified to the conversation partner. I understand heels.

Moreover, LED28 displays in the direction in which the 1st speaker 15 of the 1st directivity speaker part 13 outputs an audio | voice. For this reason, the conversation person B can recognize the state of the speech translation in the same manner as the above-described conversation person A if he / she is at a position where the voice translated into the second language output from the first speaker 15 can be heard. .
In addition, the first directional speaker unit 13 and the second directional speaker unit 24 output the voice of the translation result with directivity, so that these output voices are output to the first microphone 10 and the second microphone 21. It is possible to prevent malfunctions such as re-translating the collected sound. Further, it is possible to reduce the possibility of the translation result being heard by a third party other than the conversation partner. Moreover, it can be avoided that the speaker hears the translation result, cares about the translation result, and makes it difficult for the speaker to speak.

  FIG. 2 is a schematic external view showing the external appearance of the speech translation apparatus 1 in the present embodiment. In the present embodiment, the speech translation apparatus 1 is provided in a clamshell type mobile phone terminal. 2A is an external view of the inside of the mobile phone terminal including the speech translation device 1 in an opened state, and FIG. 2B is an external view of the outside in the open state. As shown in FIG. 2A, the second speaker 26, the display unit 17, and the first microphone 10 face the same direction, and the operation unit 20 is provided on the inner surface. Further, as shown in FIG. 2B, the LED 28, the camera 31, the first speaker 15, and the second microphone 21 face the same direction.

That is, the direction of the sound source that can be collected most efficiently by the first microphone 10 substantially coincides with the direction in which the second speaker 26 of the second directional speaker unit 24 outputs sound, and the second microphone 21 is the most efficient. The direction of the sound source that can collect sound well and the direction in which the first speaker 15 of the first directional speaker unit 13 outputs sound substantially coincide. The direction of the sound source that can collect sound most efficiently by the first microphone 10 and the second microphone 21 is the direction of the opening provided to guide the sound wave to each of these microphones. The display unit 17 displays in a direction in which the second speaker 26 outputs sound, and the LED 28 displays in a direction in which the first speaker 15 outputs sound. Furthermore, the direction in which the sound is output from the first speaker 15 and the second speaker 26 is opposite.
With such an arrangement, when the inside shown in FIG. 2A is directed to the conversation person A holding the speech translation apparatus 1, the outside shown in FIG. 2B is directed to the conversation person B who is the conversation partner. The speech translation apparatus 1 is easy to use.

  FIG. 3 is a flowchart for explaining the operation of the speech translation apparatus 1 according to this embodiment. A case where the conversation person A translates from the first language to the second language will be described as an example. First, the first instruction detection unit 19 performs a process of detecting whether a key button of the operation unit 20, for example, a “decision” key button is pressed, that is, a process of detecting a translation start instruction (S1). At this time, when the conversation person A does not press the key button of the operation unit 20 and the first instruction detection unit 19 determines that there is no instruction to start translation (S2-No), the process returns to step S1 to detect Repeat the process.

  On the other hand, when the conversation person A is pressing the key button of the operation unit 20 and the first instruction detection unit 19 determines that there is a translation start instruction in step S2 (S2-Yes), the first translation unit 11 Causes the second translation state notification unit 27 to notify the start of reception of the signal to be translated from the first microphone 10 by using the LED 28 (S3), and the speech signal to be translated from the first microphone 10 (first (S4) is started to be received (S4). The first translation unit 11 performs speech recognition processing on the received speech signal, converts it into a character string in the first language, and stores it in the first speech storage unit 12. When the first translation unit 11 detects a conversation break on the condition that a silent part longer than a predetermined time continues, the first translation unit 11 stops accepting the speech signal to be translated, and further notifies that translation is in progress. In this manner, the first translation state notification unit 16 and the second translation state notification unit 27 are instructed (S5).

  Next, the 1st translation part 11 reads the character string of the 1st language memorize | stored in the 1st audio | voice storage part 12, and performs the translation process to a 2nd language (S6). When the translation process ends, the first translation unit 11 instructs the first translation state notification unit 16 to notify that the translation result is being output (S7). Next, the 1st translation part 11 outputs the audio | voice signal of a translation result from the 1st directivity speaker part 13 (S8).

  In addition, when it is desired to confirm the translation result of what he / she spoke, the first speaker 15 adds the voice of the signal output from the second translation unit 22 in addition to the voice of the signal output from the first translation unit 11. The second speaker 26 may output the sound of the signal output from the first translation unit 11 in addition to the sound of the signal output from the second translation unit 22.

  In addition, the first translation unit 11 receives a signal resulting from translation processing of the signal output from the second microphone 21 from the second translation unit 22 and converts the voice represented by the signal into the second language. A translation process for generating a signal to be expressed is performed, a signal of the translation process result is output to the first directional speaker unit 13, and the second translation unit 22 is further output from the first translation unit 11 to the first microphone 10. A signal obtained as a result of translation processing of the received signal, a translation process is performed to generate a signal representing a voice converted from the voice represented by the signal into the first language, and the signal resulting from the translation process is sent to the second directional speaker. You may make it output to the part 24. FIG. As a result, the retranslation result of the translation result can be heard to know the validity of the translation.

In the present embodiment, the second translation start instruction unit 29 has been described on the assumption that the detection of a predetermined shape or the motion of the subject from the image captured by the camera 31 is an instruction to start translation. A signal output from the microphone 21 may be received, and detection that the signal satisfies a predetermined condition (for example, a predetermined word, frequency, etc.) may be used as an instruction to start translation.
In the present embodiment, when the first translation unit 11 receives a translation start instruction from the first translation start instruction unit 18, the first translation unit 11 starts receiving a signal from the first microphone 10, and then performs a translation process. As described above, the first translation unit 11 receives a signal from the first microphone 10 and stores a speech signal or a character string converted by speech recognition in the first speech storage unit 12 to start the first translation. When receiving an instruction to start translation by the instruction unit 18, a speech signal or a character string may be read from the first speech storage unit 12 and the translation process may be started. Similarly to the first translation unit 11 described above, the second translation unit 21 reads a speech signal or a character string from the second speech storage unit 23 when receiving a translation start instruction from the second translation start instruction unit 29. The translation process may be started.

[Second Embodiment]
The second embodiment of the present invention will be described below with reference to the drawings. FIG. 4 is a schematic block diagram showing the configuration of the speech translation apparatus 1a according to this embodiment. In the figure, the same reference numerals (10 to 15, 18 to 26, 29 to 31) are assigned to portions corresponding to the respective portions in FIG. The speech translation apparatus 1a includes a first microphone 10, a first translation unit 11, a first speech storage unit 12, a first directional speaker unit 13, a first translation state notification unit 16a, a first translation start instruction unit 18, and a second A microphone 21, a second translation unit 22, a second voice storage unit 23, a second directional speaker unit 24, a second translation state notification unit 27a, and a second translation start instruction unit 29 are provided.

  The speech translation device 1a does not include the display unit 17 and the LED 28, unlike the speech translation device 1 of FIG. 1, and instead of the first translation state notification unit 16 and the second translation state notification unit 27, the first translation state notification unit 16a and the second translation state notification unit 27a are different. The first translation state notification unit 16a is notified from the first translation unit 11 and the second translation unit 22, acquires their operation states, generates a sound signal representing the acquired states, and generates second directivity. Output to the control unit 25. As a result, the first translation state notification unit 16a causes the second speaker 26 to output the sound representing the acquired state with directivity. In addition, the second translation state notification unit 27a is notified from the first translation unit 11 and the second translation unit 22, acquires the operation state thereof, generates a sound signal representing the acquired state, and generates the first signal. Output to the directivity control unit 14. Thereby, the second translation state notification unit 27a causes the first speaker 15 to output the sound representing the acquired state with directivity.

As described above, the sound representing the state of the first translation unit 11 and the second translation unit 22 is output with directivity from the first speaker 15 or the second speaker 26, so that the sound in the first embodiment is output. Even if the display unit 17 and the LED 28 are not provided as in the translation device 1, the user using the speech translation device 1a can grasp these states.
In the first embodiment, the first translation state notification unit 16 displays the operation state of the first translation unit 11 and the second translation unit 22 on the display unit 17 to notify the conversation person A, and the second translation state The notification unit 27 turns on the LED 28 to notify the conversation person B. In the second embodiment, the first translation state notification part 16a outputs the voice to the second speaker 26 to notify the conversation person A, and the second translation. It has been described that the state notification unit 27a outputs a sound to the first speaker 15 to notify the conversation person B. However, the notification by the first translation state notification units 16 and 16a only needs to be performed in the direction in which the second directional speaker unit 24 outputs sound, and the notification by the second translation state notification units 27 and 27a is The direction of notification is not limited to the above-described form, as long as it is performed in the direction in which the unidirectional speaker unit 13 outputs sound.

  1, the first translation unit 11, the first directivity control unit 14, the first translation state notification unit 16, the first instruction detection unit 19, the second translation unit 22, the second directivity control unit 25, the first The second translation status notification unit 27, the second instruction detection unit 30, and the first translation status notification unit 16a and the second translation status notification unit 27a in FIG. 4 may be realized by dedicated hardware. These units may be constituted by a memory and a CPU (central processing unit), and the functions may be realized by loading a program for realizing the functions of the units into the memory and executing the programs.

  In addition, the first translation unit 11, the first directivity control unit 14, the first translation state notification unit 16, the first instruction detection unit 19, the second translation unit 22, the second directivity control unit 25, the second in FIG. A program for realizing the functions of the translation state notification unit 27, the second instruction detection unit 30, and the first translation state notification unit 16a and the second translation state notification unit 27a in FIG. 4 is recorded on a computer-readable recording medium. Then, the program recorded in the recording medium may be read into the computer system and executed to execute the processing of each unit. Here, the “computer system” includes an OS and hardware such as peripheral devices.

  The “computer-readable recording medium” refers to a storage device such as a flexible medium, a magneto-optical disk, a portable medium such as a ROM and a CD-ROM, and a hard disk incorporated in a computer system. Furthermore, the “computer-readable recording medium” dynamically holds a program for a short time, like a communication line when transmitting a program via a network such as the Internet or a communication line such as a telephone line. In this case, a volatile memory in a computer system serving as a server or a client in that case is also used to hold a program for a certain period of time. The program may be a program for realizing a part of the functions described above, and may be a program capable of realizing the functions described above in combination with a program already recorded in a computer system.

  The embodiment of the present invention has been described in detail with reference to the drawings. However, the specific configuration is not limited to this embodiment, and includes design changes and the like within a scope not departing from the gist of the present invention.

  The present invention is suitable for use in portable information devices such as mobile phone terminals, PDAs (Personal Digital Assistants), and electronic dictionary devices, but is not limited thereto.

It is a block diagram which shows the structure of the speech translation apparatus 1 by 1st Embodiment of this invention. It is a schematic external view which shows the external appearance of the speech translation apparatus 1 in the embodiment. It is a flowchart explaining operation | movement of the speech translation apparatus 1 in the embodiment. It is a block diagram which shows the structure of the speech translation apparatus 1a by 2nd Embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1, 1a ... Speech translation apparatus 10 ... 1st microphone 11 ... 1st translation part 12 ... 1st audio | voice storage part 13 ... 1st directivity speaker part 14 ... 1st directivity control part 15 ... 1st speaker 16, 16a ... First translation state notification unit 17 ... display unit 18 ... first translation start instruction unit 19 ... first instruction detection unit 20 ... operation unit 21 ... second microphone 22 ... second translation unit 23 ... second voice storage unit 24 ... first Two-directional speaker unit 25 ... second directivity control unit 26 ... second speaker 27, 27a ... second translation state notification unit 28 ... LED
29 ... Second translation start instruction unit 30 ... Second instruction detection unit 31 ... Camera

Claims (11)

A first microphone that collects sound in a first language, converts the collected sound into a signal, and outputs the signal;
A first translation unit which receives a signal output from the first microphone, performs a translation process for generating a signal representing a voice obtained by converting the voice represented by the signal into a second language, and outputs a signal of the translation process result; ,
A first speaker unit that outputs a sound of a signal output by the first translation unit;
A second microphone that collects sound in a second language, converts the collected sound into a signal, and outputs the signal;
A second translation unit that receives a signal output from the second microphone, performs a translation process for generating a signal representing a voice obtained by converting the voice represented by the signal into a first language, and outputs a signal of the translation process result; ,
A second speaker unit that outputs the sound of the signal output by the second translation unit;
The direction of the sound source that the first microphone can collect sound most efficiently coincides with the direction in which the second speaker unit outputs sound,
A speech translation apparatus, wherein a direction of a sound source that can be most efficiently picked up by the second microphone and a direction in which the first speaker unit outputs a sound substantially coincide with each other.   The speech translation apparatus according to claim 1, wherein the first speaker unit and the second speaker unit are directional speakers that output speech with directivity. A first translation state notifying unit for notifying that the second speaker unit is receiving a signal to be translated from the second microphone in a direction in which the second speaker unit outputs sound;
A second translation state notifying unit for notifying that the first speaker unit is receiving a signal to be translated from the first microphone in a direction in which the first speaker unit outputs a voice; The speech translation apparatus according to claim 2. A first translation state notifying unit for notifying that the first translation unit is performing a translation process in a direction in which the second speaker unit outputs audio;
The voice according to claim 2, further comprising: a second translation state notifying unit that notifies that the second speaker is performing a translation process in a direction in which the first speaker unit outputs the voice. Translation device. A first translation state notifying unit for notifying that the second speaker unit is outputting a voice that the first translation unit is outputting a speech signal of the translation processing result;
A second translation state notifying unit for notifying that the second speaker is outputting a speech signal as a result of translation processing in a direction in which the first speaker unit outputs the sound. The speech translation apparatus according to claim 2. The first speaker unit outputs the audio of the signal output by the second translation unit in addition to the audio of the signal output by the first translation unit,
3. The speech translation apparatus according to claim 2, wherein the second speaker unit outputs the speech of the signal output from the first translation unit in addition to the speech of the signal output from the second translation unit. . The first translation unit further receives a signal obtained as a result of translation processing by the second translation unit, performs a translation process for generating a signal representing a voice obtained by converting the voice represented by the signal into a second language, Output the processing result signal,
The second translation unit further receives a signal obtained as a result of the translation processing by the first translation unit, performs a translation process for generating a signal representing a voice obtained by converting the voice represented by the signal into a first language, and The speech translation apparatus according to claim 2, wherein a signal of a processing result is output. A first translation start instruction unit for receiving an instruction to start translation processing by the first translation unit;
A second translation start instruction unit for receiving an instruction to start translation processing by the second translation unit;
When the first translation start instruction unit receives an instruction to start translation, the first translation unit starts a translation process of receiving a signal output from the first microphone and converting it into a first language;
When the second translation start instruction unit receives an instruction to start translation, the second translation unit starts a translation process of receiving a signal output from the first microphone and converting it into a first language. The speech translation apparatus according to claim 1. The said 2nd translation start instruction | indication part is provided with the imaging part which images the output direction of the said 1st speaker part, and detects the said instruction | indication of the said translation start from the imaging result by this imaging part. Speech translation device. The speech translation apparatus according to claim 8, wherein the second translation start instruction unit receives a signal output from the second microphone and detects the instruction to start translation from the signal. In a speech translation method in a speech translation device,
A first process in which a first microphone of the speech translation apparatus collects speech in a first language, converts the collected speech into a signal, and outputs the signal;
The speech translation apparatus receives the signal output in the first process, performs a translation process for generating a signal representing the speech converted from the speech represented by the signal into a second language, and outputs a signal of the translation process result A second process of outputting
The speech translation device outputs the speech of the signal output in the second process in a direction substantially coincident with the direction of the sound source that the second microphone of the speech translation device can collect most efficiently. Process,
A fourth step in which the second microphone of the speech translation apparatus collects the speech in the second language, converts the collected speech into a signal, and outputs the signal;
The speech translation apparatus receives the signal output in the fourth process, performs a translation process for generating a signal representing the speech converted from the speech represented by the signal into the first language, and outputs a signal of the translation process result A fifth process of outputting
A sixth process in which the speech translation apparatus outputs the speech of the signal output in the fifth process in a direction substantially coincident with the direction of the sound source that the first microphone can collect most efficiently. A speech translation method characterized by the above.

Images