JP2007334968A - Voice switching apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a voice switching apparatus in which surrounding change can be accurately recognized when a user listens to music. <P>SOLUTION: The apparatus is provided with: a storage device 101 storing external sound as a recorded sound; an analysis device 103 discriminating whether the external sound satisfies discrimination conditions or not; a reproducing device 102 reproducing recorded sound stored in the storage device 101; and a switching unit 105 switching alternatively music input from a music reproducing device 10, the external sound, and the recorded sound reproduced by the reproducing device, and outputting it to a loudspeaker 31, and when it is decided that the external sound satisfies discrimination conditions, the sound recorded a predetermined time before the satisfaction of conditions is detected is reproduced and output from the loudspeaker 31. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a voice switching device, and more particularly, to a voice switching device that allows a user to accurately grasp a necessary surrounding situation while listening to music with headphones or the like.

  Conventionally, portable portable music playback apparatuses that play back music data stored in a recording medium such as a CD, MD, or flash memory are known. The portable music player usually listens to music played only by the user using headphones. When playing music from a portable music player and listening with headphones, it is difficult to hear external sounds. This is a good point for listening to music only. For example, when using a public transportation facility such as a train, you may not be able to hear the broadcasts on the train (such as broadcasts that inform you of the stop station). There is a problem that there is. Moreover, when carrying out a conversation, the headphones must be removed one by one, which is problematic.

  In order to solve the above problem, for example, in the portable audio reproduction device of Patent Document 1, external audio input from a microphone is recognized as voice, and the recognized voice is matched with a predetermined phrase stored in advance. Discloses a technique for notifying that effect.

  However, in Patent Document 1, for example, when the target disembarkation station is registered as “Urawa”, it is determined that the announcements “Minami Urawa” and “Kita Urawa” match as a result of voice recognition. , Because the user will be notified of "Urawa", the problem is that the user cannot determine whether "Urawa" is "Minami Urawa" or "Kita Urawa". is there.

  Further, in Patent Document 1, for example, when the user is getting off at “Urawa” station, if the surrounding person utters “Urawa” at a place different from the destination, it is determined as a result of voice recognition. Then, the user may be notified of “Urawa”. However, there is a problem in that it cannot be immediately determined whether the notification is due to a conductor announcement or a surrounding person.

  In addition, there is a case where the user uses the music playback because “the ambient noise is not preferable, so the music playback is used”. However, in Patent Document 1, the user wants to stop the music playback when the ambient noise becomes quiet. There is a problem that the request cannot be met.

Moreover, in the headset of patent document 2, the external sound input from a microphone is recognized as voice,
As a result of voice recognition, when a specific keyword, a voice print of a specific person, or a sound of a certain reference volume or higher is detected for a certain period of time, the music output input from the music playback device is stopped and the user A technique for making a sound heard is disclosed. According to Patent Document 2, it is possible to immediately respond to a call of a specific person or a call of a telephone even while listening to music.

  However, in Patent Document 2, when the music being played is muted by “voice pattern matching” and “detection of sound above a certain reference volume”, it is determined what kind of call and music reproduction is muted by the sound. There is a problem that can not be.

  For example, if the voiceprint matches a call to a user who is listening to music saying “Let's go for dinner,” the music playback was muted, but the user was muted. There is a problem of not knowing. For example, if a door buzzer sounds in the vicinity of a user listening to music, the music playback is muted, but it cannot be determined whether the interruption is a door buzzer, a telephone call, or someone's call. There's a problem. Furthermore, as in Patent Document 1, there is a problem that it is impossible to respond to a request to stop music playback when ambient noise becomes quiet.

JP 2001-256771 A Japanese Patent Laid-Open No. 2005-192004

  The present invention has been made in view of the above problems, and mainly provides a voice switching device capable of accurately grasping changes in the surrounding situation when a user is listening to music. Objective.

  In order to solve the above-described problems and achieve the object, the present invention relates to storage means for storing external sound input from a microphone as recorded sound and whether the external sound input from the microphone matches a determination condition. An analysis means for determining whether or not, a playback means for playing back the recorded voice stored in the storage means, music input from a music playback device, external sound input from the microphone, and playback by the playback means Switching means for selectively switching the recorded sound and outputting it to a speaker, and when the external sound input from the microphone matches the determination condition by the analysis means, the reproduction means The recorded sound from a predetermined time before the point in time when the sound is detected is reproduced, and the switching means outputs the reproduced recorded sound to a speaker.

  Exemplary embodiments of an audio switching device according to the present invention will be explained below in detail with reference to the accompanying drawings. The present invention is not limited to the embodiments. In addition, constituent elements in the following embodiments include those that can be easily assumed by those skilled in the art or that are substantially the same.

(Embodiment)
The audio switching device according to the present embodiment includes a storage unit that stores external audio input from a microphone as recorded audio, and an analysis unit that determines whether the external audio input from the microphone matches a determination condition. And playback means for playing back the recorded voice stored in the storage means, music input from the music playback device, external voice input from the microphone, and recorded voice played back by the playback means. Switching means for switching and outputting to a speaker, and when the external sound input from the microphone is determined to match the determination condition by the analysis means, the playback means is more predetermined than when the condition match is detected. The recorded voice from before the time is played back, and the switching means outputs the played back recorded voice to the speaker. The predetermined time is desirably a length including the voice that matches the determination condition and the preceding time required for the user to grasp the context of the matched voice (see ΔT = Δt1 + Δt3 in FIG. 6).

  According to the audio switching device of the present embodiment, when the external audio input from the microphone matches the determination condition, the recorded audio from a predetermined time before the current time can be reproduced and output to the speaker. When listening to music using headphones or the like, it becomes possible to accurately grasp changes in the surrounding environment.

Further, the playback means plays back the recorded voice in a shortened manner (plays back in a time shorter than the recording time of the recorded voice), specifically, plays back at a playback speed faster than normal voice playback (speed playback), or It is desirable to reproduce the silence (SKIP reproduction) by skipping the silent part.
Further, it is desirable that the reproduction means performs reproduction of the recorded voice until it catches up with real time. As a result, it is possible to catch up with live external audio and to smoothly switch consciousness.

  In addition, it is desirable to provide a sound reproducing means for reproducing a predetermined sound before reproducing the recorded sound. As a result, the user can know in advance that the recorded voice will be reproduced.

  Further, the analysis means stores in advance one or more of keywords, voiceprints, voice patterns, and voice levels as determination conditions, and external voices input from the microphone are keywords, voiceprints, voice patterns, and voices. It is desirable to determine whether or not the level matches. As a result, when the external sound matches the keyword, voiceprint, sound pattern, and sound level conditions, the user can listen to the recorded sound from a predetermined time before the current time. Train users who are watching can accurately identify external announcements such as getting off stations, and can recognize external calls to users and changes in the surrounding environment. Become.

  Further, the playback means plays back the recorded voice in a shortened manner (plays back in a time shorter than the recording time of the recorded voice), specifically, plays back at a playback speed faster than normal voice playback (speed playback), or It is desirable to reproduce the silence (SKIP reproduction) by skipping the silent part. As a result, it is possible to catch up with live external audio and to smoothly switch consciousness.

  Further, the analysis means stores in advance one or more of keywords, voiceprints, voice patterns, and voice levels as determination conditions, and external voices input from the microphone are keywords, voiceprints, voice patterns, and voices. It is desirable to determine whether or not the level matches. Thus, when the external sound matches the keyword, voiceprint, sound pattern, and sound level conditions, the user can record the sound from a predetermined time before the current time (ΔT = Δt1 + Δt3 in FIG. 6). For example, train users who are listening to music using headphones can accurately identify external announcements such as getting off stations, and can also call users from outside. It becomes possible to recognize changes in the surrounding environment.

  The speaker is preferably a speaker built in the headphones. This makes it possible to accurately grasp changes in the surrounding situation while listening to music played by the audio playback device using headphones.

(Example)
FIG. 1 is a diagram showing an external configuration of a headphone system to which an audio switching device according to the present invention is applied. In the figure, 10 is a portable audio playback device that plays back music data stored in a recording medium such as a CD, MD, or flash memory, 20 is a switch unit on which an operation unit 21 and a microphone 22 are mounted, 21 Is an operation unit for operating music playback / stop of the audio playback device 10, 22 is a microphone for collecting external audio, 30 is a headphone, 31 is a speaker mounted on the headphone 30. Show. The voice switching device 100 according to the present invention is mounted inside the switch unit 20.

  FIG. 2 is a diagram showing a block configuration of the voice switching device 100 according to the present invention. As shown in FIG. 2, the voice switching device 100 includes a storage device 101, a playback device 102, an analysis device 103, a detector 104, and a switching device 105.

  The microphone 21 collects surrounding sounds from the beginning to the end while the sound switching device 100 is being used (power supply is operating), and the collected sound is output as external sound to the storage device 101, the sound analysis device 103, and the switch 105. Is done.

  The switch 105 selectively selects external sound input from the microphone 21, recorded sound input from the playback device 102, or music input from the music playback device 10 in accordance with a switching instruction from the detector 104 or the operation unit 22. And output to the speaker 31.

  The storage device 101 continues recording external sound input from the microphone 21 while the power supply is operating. Here, ΔT is the time during which the recorded sound is retained by recording.

  The detector 104 detects the input / stop of music from the music playback device 10 and the input of recorded sound from the playback device 102, and outputs a switching instruction to the switch 105 and an analysis instruction to the analysis device 103.

  When the recorded sound playback instruction (PLAY) is input from the analysis device 103, the playback device 102 plays back the external sound stored in the storage device 101 as recorded sound and outputs it to the detector 104 and the switch 105.

  When the determination start instruction is input from the detector 104, the analysis apparatus 103 recognizes the external sound input from the microphone 21, and the determination condition registered in the determination condition table (see FIG. 4) stored in advance. It is determined whether or not it matches (predetermined keyword, voice print, voice pattern, voice level). If they match, the analysis is stopped and a recorded voice playback instruction (PLAY) is output to the playback device 102. .

  Specifically, the analysis device 103 performs voice recognition for voice recognition, voiceprint recognition for voiceprint recognition, voice pattern recognition for voice pattern recognition, and voice level measurement for measuring voice level with respect to input external voice. Means for detecting whether or not the input external voice matches a predetermined keyword, voice print, voice pattern, and voice level registered in the condition determination table in advance.

  FIG. 4 is a diagram illustrating a configuration example of the determination condition table. In the determination condition table, as shown in FIG. 4, predetermined keywords, voiceprints, voice patterns, and voice level comparison elements, and related words / phrases for notification are registered. The data of the determination condition table can be arbitrarily registered by the user using an input means (not shown). For example, in the determination condition table shown in FIG. 4, “Urawa” as a predetermined keyword, “voice print data of friend A” as a voice print, “20 dB or less” as a voice level, and “phone bell voice pattern data” as a voice pattern. The analysis apparatus 103 is registered and determines whether or not the external voice matches at least one of these comparison elements (determination conditions). When the condition match is determined, the sound of the corresponding related phrase may be output from the speaker 31 to notify the user of the determination condition under which the match is detected. For example, in the example shown in FIG. 4, when a condition match is detected with “Urawa” of “predetermined keyword”, the fact that “Urawa” has been detected is notified, and “voice print data of friend A” is used. If a condition match is detected, a notification that “friend A's voiceprint” has been detected is sent.

  In addition, here, the determination condition is a predetermined keyword, a voiceprint, a voice pattern, and a voice level. However, it is not necessary to set all of them as the determination condition, and one of the predetermined keyword, voiceprint, voice pattern, and voice level. Alternatively, a plurality of determination conditions can be used. Further, here, the determination of coincidence is an OR condition, but an AND condition may be used.

  FIG. 3 is a flowchart for explaining the operation of the voice switching device 100 of FIG. In the following description, a case will be described in which the user uses the headphones 30 to watch music played on the music playback device 10.

  In FIG. 3, first, in the voice switching device 100, when the power is turned on, the microphone 21 collects surrounding voices and outputs them as external voices to the storage device 101, the voice analysis device 103, and the switch 105. To do. Further, the switch 105 switches the SW to the terminal a and outputs the external sound input from the microphone 21 from the speaker 31. Thus, the user wearing the headphones 30 can clearly hear the sound from the surroundings obstructed by the headphones 30 via the sound switching device 100. In other words, it is possible to check the surrounding sounds and talk in real time without removing the headphones 30. In addition, the storage device 101 continues to record external sound input from the microphone 21. The time retained by recording is ΔT.

  Subsequently, the detector 104 detects whether or not music is input from the music playback device 10 by detecting a music input signal from a voltage level or the like (step S11). (“Yes” in step S11), external voice analysis processing is executed (step S12). In the present embodiment, the audio switching device 100 is configured to be able to operate autonomously by detecting the music input signal from the voltage level or the like. However, the start of music playback is detected from the music playback device 10. The determination may be made by receiving the instruction as a control signal. In addition, a beep sound or a sound effect may be generated at the timing of voice switching. As a result, the user can be made aware that a selection voice different from the previous one will be output. In addition, in order to inform the user of the current switching status, a switch device 20 is provided with a display device such as an LED or a liquid crystal panel to notify the switching status (what audio is being output). May be.

  Here, the external audio analysis process will be described in detail. When detecting the music input, the detector 104 outputs a switching instruction (SW: b) for performing music output to the switching unit 105 and at the same time a determination start instruction (analysis instruction) to the analysis device 103. Is output.

  When the switching instruction (SW: b) for performing music output from the detector 104 is input to the switching device 105, the switching device 105 switches the SW to the terminal b, and the audio output is output to the speaker 31. Thereby, the user can listen to the music reproduced from the speaker 31. When the analysis apparatus 103 receives a determination start instruction from the detector 104, the determination condition (predetermined keyword, voiceprint, sound pattern, sound level) registered in the determination condition table matches the input external sound. In step S15, if not matched, the process returns to step S13. If matched, the external sound analysis process is stopped, and the process proceeds to step S16.

  While detecting the music reproduction, the detector 104 also monitors the music stop (step S13). When detecting the music stop, the detector 104 performs an analysis stop process (step S14), and then returns to step S11. The detection of music stop may be determined by receiving a stop instruction from the external music playback device 10 as a control signal.

  In this analysis stop processing, when detecting the music stop, the detector 104 outputs a switching instruction (SW: a) for outputting an external sound to the switch 105 and determines to the analysis device 103. Outputs a stop instruction (analysis stop instruction). When a switching instruction (SW: a) for outputting an external sound is input from the detector 104, the switch 105 switches the SW to the terminal a, and the external sound is output to the speaker 31. As a result, the user can check the surrounding sounds and have a conversation in real time without removing the headphones 30 while the music is stopped. In addition, when the analysis apparatus 103 receives the determination stop instruction from the detector 104, the analysis apparatus 103 stops the analysis of the external sound.

  In step S16, a stored audio reproduction process is executed. In the stored voice reproduction process, the analysis apparatus 103 outputs a recorded voice reproduction instruction (PLAY) of the stored recorded voice to the reproduction apparatus 102. When the recording device playback instruction (PLAY) is input from the analysis device 103, the playback device 102 plays back the external sound stored in the storage device 101 as the recorded sound and outputs it to the detector 104.

  Note that music playback of the music playback device 10 continues during recording and playback, but when a control line is provided between the music playback device 10 and the voice switching device 100, the playback is synchronized with the recording and playback. Thus, the music playback stop signal may be notified to the external music playback device 10. As a result, it is possible to avoid a situation in which music continues to be played back in situations not intended by the user.

  When the detector 104 detects the recorded sound output from the playback device 102, the detector 104 outputs a switching instruction (SW: c) for outputting the recorded sound to the speaker 31 to the switch 105. When the switching instruction (SW: c) for performing the recording sound output from the detector 104 is input from the detector 104, the switching device 105 switches the SW to the terminal c, and the recording sound is output to the speaker 31.

  Here, the recorded sound reproduced by the reproducing apparatus 102 will be described. The reproduction of the recorded voice by the playback device 102 starts from the past of the predetermined time ΔT from the recording voice start time (when the coincidence of the determination conditions is detected). FIG. 5 is a diagram for explaining a recorded sound reproduction method (part 1). The predetermined time ΔT desirably satisfies the following conditional expression (1).

Predetermined time ΔT = predetermined sound output time Δt1 + predetermined time Δt3 (1)
In FIG. 5, Δt1 is a time during which a predetermined sound is output, and for the sake of convenience of explanation, it is assumed that it is the same time as the analysis time necessary for determining whether the analysis apparatus 103 matches the condition. For example, Δt1 is a time during which “Urawa” sound is output when “Urawa” is registered as the predetermined keyword. Δt3 is a preceding time required for the user to grasp the context of the matched voice. For example, when Δt2 is a time for outputting “Minami” as a sound, Δt3 is a time that can sufficiently determine “Minami”. Note that Δt3 may be arbitrarily set by the operator.

  In this way, by playing back the recorded voice from the past for a predetermined time ΔT for the keyword “Urawa”, it is possible to grasp the context, and whether the judgment match is purely “Urawa”, “ It is possible to determine whether it has been hit with Minami Urawa or Kita Urawa. Also, by playing the external voice directly as recorded voice for the keyword “Urawa”, whether the matching situation is due to the conductor announcement for getting off the vehicle, or the surrounding people talks regardless of the destination. Can be determined. As a result, the user can accurately grasp the surrounding conditions necessary for the user while listening to music with the headphones 30.

  Also, in the case of [voice print data], for example, when a friend A calls “Let's go to eat dinner”, the situation of the determination coincidence [voice print of friend A] is expressed as ΔT from the past. By playing, it is possible to determine what the friend A called when calling.

  Also, in the case of [Voice Pattern], even if the telephone bell rings, and even if the telephone bell stops ringing immediately after that, the situation where the judgment coincides can be reproduced by playing the recorded voice from the past ΔT. , You can determine that the phone bell rang.

  Further, in the case of [audio level], it is possible to determine a situation where the surrounding situation is quiet. This is suitable for a person who cares about sound leakage in the surrounding situation because the surroundings have become quieter and the volume of music is reduced. Note that it may be determined that the surroundings are noisy and the music volume is increased because the surroundings are noisy.

  Note that, as described above, the speaker 31 may be notified to the user of what caused the current determination immediately before the reproduction of the recorded sound is started. This is because this notification allows the user to acquire superior data by grasping the surrounding situation. Or you may make it call attention with a chime sound. These may be used in combination.

  In addition, when the recorded voice is played back, the playback is shortened. Here, the shortened playback means playback in a shorter time than the recording time of the recorded voice, for example, playback at a playback speed faster than normal voice playback (speed playback), or efficient silence portion. SKIP playback (SKIP playback).

  FIG. 6 is a diagram for explaining a method (part 2) for reproducing a recorded sound according to the present embodiment. In FIG. 6, when the recorded sound is reproduced at the normal speed, the sound output from the speaker 31 is always the past sound by ΔT. In this embodiment, the difference in ΔT is reproduced faster than the normal speed, or the silence is skipped so that the difference approaches “0”, and the shortened reproduction time (ΔS) has elapsed. At the time, that is, when the difference becomes “0”, the playback of the recorded sound by the playback device 102 is stopped, and the switcher 105 switches the speaker output to the external sound. As a result, the target external sound can be recognized while listening to music with the headphones 30, and the return (recognition switching) after the external sound recognition can be performed smoothly.

  For example, assuming a situation in which a friend A talks to music while listening to music, first, “voice comprehension, friend A” is notified of a condition match by determination match by voice print. Thereby, it can be recognized that the friend A called while listening to the music. Next, by shortening and reproducing the recorded voice, it is possible to grasp what kind of content the friend A called. Furthermore, even if the friend A continues the conversation, the conversation can be heard and understood without interruption without interrupting the conversation of the friend A.

  Now, after the stored audio reproduction process in step S16 is completed, the detector 105 outputs a switching instruction (SW: a) for outputting the external audio to the speaker 31 to the switch 105. When a switching instruction (SW: a) for outputting an external sound is input from the detector 104, the switch 105 switches the SW to the terminal a and the external sound is output from the speaker 31. In this case, when the user newly listens to music, a switching instruction (SW: b) for manually outputting music to the switching unit 105 is output from the operation unit 22. In addition, it is good also as a structure which outputs the switch instruction | indication (SW: b) for outputting music to the switch 105 automatically after fixed time progress.

  As described above, according to the voice switching device 100 according to the present embodiment, the storage device 101 that stores the external voice input from the microphone 21 as the recorded voice and the external voice input from the microphone 21 are used as the determination conditions. Analyzing device 103 that determines whether or not they match, playback device 102 that plays back the recorded voice stored in the storage device 101, and music input from the music playback device 10 according to instructions from the detector 104, microphone 21 And a switch 105 that selectively switches between the external sound input from the recording device and the recorded sound reproduced by the playback device 102 and outputs the same to the speaker 31, and the external sound input from the microphone 21 by the analysis device 102 When it is determined that the determination condition is met, the recorded sound from a predetermined time before the time point when the condition match is detected is played and the recorded sound is output to the speaker 31. Therefore, when the external sound input from the microphone matches the judgment condition, the recorded sound from a predetermined time before the point of detection of the condition match can be played and output to the speaker, such as headphones. When listening to music using the, it is possible to accurately grasp the changes in the surrounding situation.

  Further, according to the audio switching device 100 according to the present embodiment, the reproducing device 102 reproduces the recorded audio in a shortened manner (reproduced in a time shorter than the recording time of the recorded audio), and more specifically, than normal audio reproduction. Since the playback is performed at a fast playback speed (speed playback) or the silent part is played back by SKIP (SKIP playback), it is possible to catch up with the live external sound and to smoothly switch the consciousness. .

  Further, according to the voice switching device 100 according to the present embodiment, the analysis device 103 stores in advance one or more of a keyword, a voiceprint, a voice pattern, and a voice level as determination conditions. When it is determined whether the input external voice matches the keyword, voiceprint, voice pattern, and voice level, etc., the external voice matches the keyword, voiceprint, voice pattern, and voice level conditions In addition, the user can listen to the recorded sound from a predetermined time ago. For example, a train user who listens to music using headphones can accurately determine an external announcement such as an exit station. It is also possible to recognize external calls to users and changes in the surrounding environment.

  Further, according to the audio switching device 100 according to the present embodiment, since the speaker 31 is built in the headphones 30, when listening to music reproduced by the audio reproducing device using the headphones. In addition, it becomes possible to accurately grasp the surrounding situation changes.

(Modification 1)
FIG. 7 is a view for explaining a first modification of the voice switching device according to the present invention. In FIG. 7, the modification 1 is different from the above embodiment in that the audio switching device 100 is mounted inside the audio reproducing device 10.

(Modification 2)
FIG. 8 is a diagram for explaining a second modification of the voice switching device according to the present invention. In FIG. 8, the modification 3 is the structure which mounted the audio | voice switching apparatus 100 inside the headphones.

(Modification 3)
In the voice switching device 100 shown in FIG. 2, a signal line (wired) is used to transmit input / output voices and control signals. However, the signal transmission is not limited to wired communication, such as Bluetooth. The wireless communication may be used.

(Modification 4)
When a conference call or video conference is performed using a headset, there may be a case where a change in the surrounding voice situation, for example, a voice call from other people or a phone call is not noticed. Therefore, in the case of conducting a telephone conference / TV conference using the headphone system according to the present embodiment, instead of inputting music from the music playback device 10, the conference audio is used as an input. Since it is difficult to determine in advance whether the audio output of the conference audio or the external audio is prioritized, the conference audio is superimposed on the conference audio and output to the speaker 31 without muting the conference audio. You may decide to do it. In addition, a display device may be used to notify the user of a situation change necessary for the user. Further, both notification by voice of situation change and notification by video may be performed, and information on what kind of situation change has occurred in voice and video may be included.

(Modification 5)
Moreover, the audio switching device 100 according to the present invention is not limited to the case where headphones are used, but may be output to indoor / in-vehicle speakers or the like. For example, a microphone is arranged at the door of the entrance, and the audio switching device 100, the music playback device, and the speaker are arranged in the room, and the sound output from the speaker is output according to the external sound input from the entrance microphone. It may be switched. Thereby, for example, when listening to music indoors, when the voice of the visitor is detected by the front microphone, the output of the speaker can be switched to the voice of the visitor.

  The audio switching device according to the present invention is useful for accurately grasping an external situation change when listening to music, and particularly useful for listening to music using headphones.

It is a figure which shows the external appearance structure of the headphone system to which the audio | voice switching apparatus which concerns on this invention is applied. It is a figure which shows the block configuration of the audio | voice switching apparatus based on this invention. It is a flowchart for demonstrating operation | movement of a voice switching apparatus. It is a figure which shows the structural example of a determination condition table. It is a figure for demonstrating the reproduction method (the 1) of recorded audio | voice. It is a figure for demonstrating the reproduction method (the 2) of recorded audio | voice. It is a figure for demonstrating the modification 1. FIG. It is a figure for demonstrating the modification 2. FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Audio | voice reproduction apparatus 20 Switch part 21 Operation part 22 Microphone 30 Headphone 31 Speaker 100 Audio | voice switching apparatus 101 Memory | storage device 102 Playback apparatus 103 Analysis apparatus 104 Detector 105 Switcher

Claims (7)

Storage means for storing external sound input from a microphone as recorded sound;
Analysis means for determining whether or not external sound input from the microphone matches a determination condition;
Playback means for playing back the recorded voice stored in the storage means;
Switching means for selectively switching between music input from a music playback device, external sound input from the microphone, and recorded sound played back by the playback means, and outputting to a speaker;
With
When it is determined by the analysis means that the external sound input from the microphone matches the determination condition, the reproduction means reproduces the recorded sound from a predetermined time before the time point when the condition match is detected, and the switching means Outputs the reproduced recording sound to the speaker.   The voice according to claim 1, wherein the predetermined time is a length including a voice that matches the determination condition and a preceding time required for a user to grasp a context of the matched voice. Switching device.   The voice switching device according to claim 1 or 2, wherein the playback means plays back the recorded voice in a shortened manner.   4. The voice switching device according to claim 3, wherein the playback means plays back the recorded voice until it catches up with real time.   The voice switching device according to any one of claims 1 to 4, further comprising voice playback means for playing back a predetermined voice before playing the recorded voice.   The voice switching device according to claim 1, wherein the determination condition is one or more of a keyword, a voiceprint, a voice pattern, and a voice level.   The voice switching device according to claim 1, wherein the speaker is a speaker built in a headphone.

Images