JP2006318182A - Sound retrieval device, sound retrieval method, sound reproduction method, portable reproduction device, musical piece retrieval device, musical piece retrieval method and musical piece reproduction method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a portable reproduction device capable of retrieving sound or a musical piece desired by a user from a plurality of pieces of sound data or musical piece data and reproducing it by easy operation without receiving attention of people around. <P>SOLUTION: This portable reproduction device reproducing the sound or the musical piece selected from a plurality of pieces of sound or a plurality of musical pieces has: a movement data input part detecting and inputting movement of the portable reproduction device; a sound information storage part storing sound information related to the plurality of pieces of the sound and the plurality of musical pieces; a decision part comparing movement data inputted in the movement data input part, and the musical piece data or the sound data stored in the sound information storage part, and outputting the sound or the musical piece similar to the movement data as a retrieval result; a display part displaying metadata of the retrieval result outputted by the decision part; and a reproduction part reproducing the musical piece data or the sound data of the retrieval result outputted by the decision part. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a voice search device, a voice search method, a voice playback method, a portable playback device, a music search device, a music search method, and a music playback method, and in particular, a plurality of voices and similar voices desired by a user from music. And technology for searching and playing music.

In recent years, with the advancement of music data compression technology and the increase in storage capacity of storage media, the number of music data that can be recorded on a recording medium has increased, and even a portable player has more than 10,000 songs. Data can be stored, and a desired music piece can be selected and enjoyed.
In such a portable player, when the user selects a desired song, the song name or artist name of the recorded song is displayed on the display device, and the selected song is actually played and confirmed. ing.

However, as described above, when the number of songs that can be selected is several hundred to 10,000, it is practically impossible to select a desired song by such a music selection method.
This can be searched by referring to the metadata if the song name, artist name, etc. are known, but cannot be searched when it is unknown.

On the other hand, when the user selects a song, the user may search for a specific song as described above, but may search for a song according to the user's preference. For example, when listening to music that matches a specific rhythm.
In this case, you will remember the song you know and confirm it by listening to the artist name. This requires considerable time and effort to search for the desired music.

In order to solve such a problem, in the technique of Patent Document 1, first, in the user terminal device, for each piece of music, music data and attribute information to be used for search are obtained via a communication line network. Alternatively, a music database is created by extracting from music data by voice recognition processing or pitch extraction processing and recorded on a recording medium. In the mobile terminal device, music is selected with reference to attribute information in the music database recorded on the recording medium in response to a user input, and the music is output to the headphones and reproduced.
Thereby, the music according to a user's feeling and a situation can be selected easily and can be enjoyed.

  On the other hand, a huge number of songs are stocked in the karaoke device, and the user can select a song to be sung by specifying the number of the song to be sung with the remote control device while referring to the table of contents. A song selection is being made. Also in this case, when the information about the music such as the music title or artist name is unknown, it cannot be searched.

  For this purpose, for example, a melody to be selected is input by using a MIDI (Musical Instrument Digital Interface) keyboard or voice, and the page is searched in the table of contents etc. by searching for a matching or similar song compared to the song data. The music can be selected without turning (see Patent Documents 2 and 3).

  In addition, there are times when you want to know the names of songs you are listening to at an outdoor live concert, the names of songs that are being broadcast on a TV song program or on the radio, or the names of songs you have met in the city. In this case, all or part of the song can be recorded, and a song name or the like can be searched based on the recorded data (see Patent Document 4).

  By the way, according to the technique of Patent Document 5, a mobile phone equipped with an acceleration sensor, a magnetic sensor, a light emitting means, and the like is swung like a penlight or a baton, so that it depends on the swinging mode detected by the sensor. In addition, it is possible to control the tempo of automatic performance, sound quality, volume, etc., and to control the light emission of the light emitting means. As a result, the automatic performance of the automatic performance device can be controlled based on the movement mode and posture state of the mobile phone generally owned by young people. Can be lowered.

Further, there is provided a mobile phone device that detects a shake state of the mobile phone device main body with a biaxial geomagnetic sensor, converts the shake state into tempo data, and changes or edits the playback state (Patent Document 6). reference).
JP 2002-49626 A JP 09-258729 A JP 09-293083 A JP 2002-215632 A JP 2001-333156 A JP 2003-108130 A

  However, in the technique of Patent Document 1, song data must be analyzed and recorded as attribute information for possible moods and situations, and in the case of attribute information that has not been analyzed, a desired song is searched. I can't.

In the techniques of Patent Documents 2 and 3, specialized techniques are required to input a melody with a MIDI keyboard, and such a device must be connected. In addition, when inputting a melody by voice, if the melody is suddenly spoken in a public place, it not only attracts the attention of the people in the vicinity, but also inconveniences the surrounding people.
Moreover, the technique of patent document 4 requires the same data, even if it is a part of music as data for search, and is not the technique for searching music according to a user preference.

The present invention has been made in consideration of the actual circumstances as described above, and is capable of searching a user's desired voice or music from a plurality of voices or music by a simple operation without attracting the attention of those around. An object is to provide a search device, a voice search method, a music search device, and a music search method.
It is another object of the present invention to provide a portable playback device, a voice playback method, and a music playback method that can play back the searched voice and music using such a voice search device and music search device.

In order to solve the above problems, the present invention has the following configuration.
The voice search device includes a determination unit that compares motion data of the device with a plurality of voice data and outputs voice data similar to the motion data as a search result.
Alternatively, a feature amount extraction unit that extracts a feature amount from motion data of the device, and a feature amount extracted by the feature amount extraction unit and a feature amount of a plurality of pieces of sound data are compared, and a sound similar to the motion data A determination unit that outputs data as a search result.

The determination unit calculates the similarity between the motion data and the audio data, and outputs a song having a similarity greater than a predetermined threshold as a search result. Here, the threshold value may be changeable.
In addition, the music search device uses music data instead of voice data in the above-described voice search device.

  The portable playback device is a portable playback device that plays back audio data selected from a plurality of audio data, and includes a motion data input unit that detects and inputs movement of the portable playback device, and sound information related to the plurality of audio data. A sound information storage unit for storing metadata, the voice search device for searching for voice data in the sound information storage unit similar to the input motion data, and a playback unit for playing back the voice data of the search result And a display unit for displaying a list of metadata relating to the audio data of the search result.

  The reproduction unit can reproduce audio data having the highest similarity among the search results, or the motion data input by the motion data input unit is data for search, or a search result selection instruction If the motion pattern determination unit determines that the search result is selected as a search result selection instruction, the audio data that is already being reproduced from the search result is provided. Another audio data may be reproduced.

Here, when the audio data selected by the reproduction unit is reproduced, metadata about the audio data is displayed, the threshold of the voice search device is set from a portable reproduction device, or the sound information The storage unit may be made detachable from the portable playback device.
Furthermore, in the portable playback device described above, music data may be used instead of audio data.

According to the present invention, since the search is performed based on the movement of the device, it is possible to search the user's desired voice or music from a plurality of voices or music by a simple operation without attracting the attention of those around.
For example, by a simple operation in which the user shakes the apparatus quickly (violently) or slowly (slowly), an up-tempo (violent rhythm) sound or music, or a slow tempo (relaxed rhythm) sound or music It can be. In addition, since the device is simply shaken, no sound leaks to the surroundings, and the surrounding people are not attracting attention and are not inconvenienced even in public places.
In this way, the user can easily select a desired voice or music piece by a simple operation.

Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings.
In the present embodiment, a case where a user searches for and plays a piece of music desired from a plurality of pieces of music will be described as an example. However, the same effect can be achieved by using voice instead of the piece of music.
In the case of voice, a voice search device is used instead of the music search device described below, and the voice information and metadata related to the voice are stored in the sound information storage unit, so that the user can select from a plurality of voice data. Can search and reproduce desired audio data.

<Embodiment 1>

  FIG. 1 is a block diagram showing a functional configuration of a music search device according to Embodiment 1 of the present invention, in which a music search device 10 includes a determination unit 11, a connector 1 (12), and a connector 2 (13). , Connector 3 (14), determination condition setting memory 15, and work memory 16.

The connector 1 (12) is for connecting to an acceleration sensor, a gyro sensor, or the like provided outside the music search device 10, and inputs the captured motion data to the determination unit 11.
The connector 2 (13) is for connecting to the sound information storage unit 20 provided outside the music search device 10, and determines the sound information of the music to be searched stored in the sound information storage unit 20. Input to section 11.
The connector 3 (14) is for connecting to the outside of the music search apparatus 10, and outputs the result determined by the determination unit 11 to the outside.

The determination condition setting memory 15 is configured by a nonvolatile memory such as an EEPROM (Electrically Erasable and Programmable ROM) or a flash memory, and stores a threshold value when it is determined that the comparison results are similar. The threshold value stored here may be fixedly stored or may be set from the outside.
The work memory 16 is configured by a RAM (Random Access Memory), and temporarily stores data processed by the determination unit 11.

When the motion data is input via the connector 1 (12), the determination unit 11 reads the music data for each music stored in the sound information storage unit 20 via the connector 2 (13), and Calculate similarity. When the calculated similarity is larger than the threshold value stored in the determination condition setting memory 15 in advance, the position information in the sound information storage unit 20 of the music and the similarity are associated and stored in the work memory 16. To do.
The determination unit 11 makes a determination on all the pieces of music stored in the sound information storage unit 20 and then arranges the similarities stored in the work memory 16 in descending order. The position information is output as a determination result to the outside via the connector 3 (14).

The sound information storage unit 20 stores a large number of music composed of metadata and music data for each music. The metadata includes a song name, a record company name, an artist name, position information indicating where the song is located in the sound information storage unit 20, and a song length.
The music data of the music is recorded in a digital data format such as MP3 (Moving Picture Experts Group Audio Layer 3), WMA (Windows (registered trademark) Media Audio), or ASF (Advanced Streaming Format).

  The motion data captured by an acceleration sensor, gyro sensor, or the like is data obtained by sampling and digitally converting vibration waveform data of a music search device or a device incorporating this music search device. For example, the motion data is represented as waveform data as shown in FIG.

  The song data to be compared with the motion data is for a predetermined length (for example, 3 to 5 seconds) of the beginning portion, the rust portion or the whole song of the song. When the position information of the rust portion is not recorded in the music metadata, it is obtained by analyzing the music data using a known technique (for example, Japanese Patent Application Laid-Open No. 2004-233965).

  The determination unit 11 compares the motion data and the music data by a known technique such as the DP matching method and calculates the similarity. For example, the motion data waveform shown in FIG. 2 and the music data waveform shown in FIG. 3 are compared by the DP matching method.

<Embodiment 2>

  FIG. 4 is a block diagram showing a functional configuration of the music search apparatus according to the second embodiment of the present invention. In the figure, the music search apparatus 10 includes a determination unit 11, a connector 1 (12), and a connector 2 (13). , Connector 3 (14), determination condition setting memory 15, work memory 16, and feature amount extraction unit 17. The same components as those in FIG. 1 are denoted by the same reference numerals, and only differences will be described.

In the second embodiment, the motion data input via the connector 1 (12) is passed to the feature amount extraction unit 17.
The feature quantity extraction unit 17 extracts a feature quantity from the waveform data of the motion data and generates a feature vector. Here, the feature amount includes, for example, the tempo, the height of the waveform, the length of the amplitude, the slowness of the rhythm, the strength, and the like, and these extraction methods can use known techniques.
The determination unit 11 extracts the feature vector of the motion data generated by the feature amount extraction unit 17 and the feature extracted from the beginning portion, the rust portion, or the whole of the song stored in the metadata of each song in the sound information storage unit 20. Similarity search is performed by comparing the vectors, and the position information of the music is output to the outside in descending order of similarity.

  As described above, if the search is performed using the feature amount of the motion data, the more music is saved, the more the comparison time is saved as compared with the first embodiment.

<Embodiment 3>

Next, an embodiment of a portable playback device using the music search device described in the above embodiment will be described.
When a user selects a desired piece of music from a plurality of pieces of music data stored in the portable playback device, the user selects a desired piece of music by slowly or vigorously shaking the portable playback device after pressing a music selection button provided in the portable playback device. The rhythm of the music is expressed (Fig. 5). The portable playback device searches for music similar to this rhythm and displays a list of music metadata (song name, artist name, etc.) on the display device.
A desired music piece is selected from the displayed contents and played back.

  FIG. 6 is a block diagram illustrating a functional configuration of the portable playback device according to the third embodiment of the present invention, in which the portable playback device includes a music search device 10, a sound information storage unit 20, a sensor 30, and a control unit. 40, a setting unit 41, a mode setting memory 42, a motion data input unit 43, a display unit 44, an input unit 45, and a playback unit 46.

The sound information storage unit 20 is configured by an HDD (Hard Disk Drive), a compact flash (registered trademark) memory, a memory stick, an SD card, or the like, and stores a plurality of music data.
As described above, the music data is composed of metadata and music data for each music. The metadata includes a song name, a record company name, an artist name, position information indicating where the song is located in the sound information storage unit 20, and a feature vector representing the length of the song and the feature of the song.
The music data of the music is recorded in the format of MP3, WMA, ASF or the like converted into digital data. The sound information storage unit 20 may be fixed to the portable playback device, or may be detachable.

The sensor 30 is composed of a gyro sensor, an acceleration sensor, or the like, and detects the shake of the portable playback device in accordance with the power ON / OFF of the sensor from the control unit 40, and outputs the waveform data as motion data.
When the motion data is sent from the sensor 30, the motion data input unit 43 samples at a predetermined cycle, digitizes the motion data within a predetermined time (for example, 3 to 5 seconds), and stores the music search device 10. Send to.
This music search device 10 may be either the above-described first or second embodiment.

The control unit 40 controls the entire portable playback device.
First, when the user wants to select music based on the movement data of the portable playback device, the user presses a music selection button provided on the portable playback device.
When the music selection button is pressed, the control unit 40 turns on the sensor 30 to detect vibration. When the music selection button is pressed again, the control unit 40 turns off the sensor 30.
In addition, when the list of music position information is passed from the music search device 10, the control unit 40 refers to the sound information storage unit 20 based on the position information of the music, and stores metadata (similar to motion data) ( A list of song names and artist names is created, and the display unit 44 is activated.
When the display unit 44 receives a list of metadata (music name and artist name, etc.) of music from the control unit 40, the display unit 44 displays the list as a list on the display device.

  When a cursor movement instruction is input, the input unit 45 moves the cursor display on the music list displayed on the display device. Further, when a playback execution instruction is input to the music indicated by the cursor, the input unit 45 returns the position information and the playback execution instruction in the list where the cursor is located to the control unit 40.

When receiving the reproduction execution instruction from the input unit 45, the control unit 40 calculates the position information of the music from the position information in the list, activates the reproduction unit 46, and passes the position information of the music.
The reproduction unit 46 reads out the music data of the music indicated by the position information passed from the control unit 40 from the sound information storage unit 20, and outputs and reproduces the data using a speaker, headphones, or the like.

When the user designates a setting value, the setting unit 41 is activated via the control unit 40.
The setting unit 41 displays a setting screen for the portable playback device and the music search device on the display device, and inputs one of the following setting values.

(1) Processing Mode There are two processing modes, a display mode (third embodiment) and a reproduction mode (fourth and fifth embodiments). In the display mode, a list of similarly searched songs is displayed on the display device, and reproduction is performed by selecting a desired song from the list according to the user's designation.
In the playback mode, songs are played back in descending order of similarity or in the order specified by the user.
The setting unit 41 stores the input processing mode in a mode setting memory 42 configured from a nonvolatile memory.

(2) Search Condition As described in the previous embodiment, a threshold value for determining the similarity by the determination unit 11 of the music search device 10 is set in the determination condition setting memory 15 of the music search device 10.
The setting unit 41 may not be provided as long as the processing mode and the search condition are determined in advance.

Next, processing until the user reproduces desired music will be described using the flowchart of FIG.
First, before using the portable playback device, the processing mode is set by the user (step S1). Of course, this operation is not necessary when the processing mode is set to the display mode as a default value.
Next, when the user depresses the music selection button, this depressing is detected (step S2), and the sensor is turned on (step S3).

When the sensor 30 is turned on and the user shakes the portable playback device with slow speed, strength, etc., the motion data of the device is sampled and read within a predetermined time (for example, 3 to 5 seconds) (step S4).
The read motion data is passed to the music search device 10, and music similar to this motion data is searched (step S5).
The metadata (song name, artist name, etc.) of the searched music is extracted from the sound information storage unit 20 and displayed on the display device as a music list of search results (step S6).

  The user moves the cursor from the list displayed on the display device to find a desired music piece and inputs a reproduction instruction. When this reproduction instruction is detected (step S7), the music data is extracted from the position information of the music on which the cursor is located, and is reproduced and output to a speaker or headphones (step S8). When it is desired to instruct the reproduction of other music on the list display, the cursor is moved and the reproduction instruction is performed again.

When the user wants to end the music selection by the movement of the portable playback device, the user presses the music selection button again. When the pressing of the music selection button is detected (step S9), the power of the sensor 30 is turned off (step S10).
Thereafter, the user performs a function operation other than the movement of the portable playback device.

<Embodiment 4>

The fourth embodiment is a portable playback device when the processing mode described in the third embodiment is a playback mode.
FIG. 8 is a block diagram illustrating a functional configuration of the portable playback device according to the fourth embodiment of the present invention, in which the portable playback device includes a music search device 10, a sound information storage unit 20, a sensor 30, and a control unit. 40, a setting unit 41, a mode setting memory 42, a motion data input unit 43, a display unit 44, and a playback unit 46. In FIG. 8, the same components as those in FIG. 6 are denoted by the same reference numerals, and only the differences will be described.

When receiving the search result from the music search device 10, the control unit 40 activates the playback unit 46 and passes the position information and the playback instruction of the music with the highest similarity among the search results.
The reproduction unit 46 extracts the music data of the music corresponding to the position information received from the control unit 40 from the sound information storage unit 20, and reproduces and outputs the music data to a speaker or headphones.

  In addition, the control unit 40 activates the reproduction unit 46, extracts metadata (music name, artist name, etc.) of the music corresponding to the position information of the music with the highest similarity in the search results, and displays the display unit 44. Also starts and passes metadata (song name, artist name, etc.). The display unit 44 displays the received metadata (song name, artist name, etc.) on the display device, thereby knowing the metadata (song name, artist name, etc.) of the music being played.

  When the search result is composed of a plurality of music pieces, it may be played back in descending order of similarity as described above, or may be played back randomly.

<Embodiment 5>

The fifth embodiment is another embodiment of the reproduction mode described in the fourth embodiment.
FIG. 9 is a block diagram illustrating a functional configuration of a portable playback device according to the fifth embodiment of the present invention, in which the portable playback device includes a music search device 10, a sound information storage unit 20, a sensor 30, and a control unit. 40, a setting unit 41, a mode setting memory 42, a motion data input unit 43, a display unit 44, a playback unit 46, and a motion pattern determination unit 47. 9, the same components as those in FIG. 8 are denoted by the same reference numerals, and only the differences will be described.

In the fifth embodiment, when a plurality of music pieces similar to the motion data are searched, the user selects a music piece to be played back.
When searching for music, for example, it is determined to shake the portable playback device in the vertical direction for 3 to 5 seconds or more. Also, a swing within 3 seconds is not adopted as motion data for music search.
In order to select a music piece from a plurality of search results, for example, it is decided to shake the portable playback device within 3 seconds in the right or left direction. In addition, a swing of 3 seconds or more is adopted as motion data for music search. For the swing to the right, move the cursor to the smaller similarity and select and play the music. For the swing to the left, move the cursor to the higher similarity and select and play the music.

When the swing of the portable playback device is determined as described above, the motion pattern determination unit 47 searches the captured motion data for music if the motion data captured by the motion data input unit 43 continues for 3 seconds or longer. It is passed to the device 10 to search for similar music.
In addition, the movement pattern determination unit 47 sends a right direction instruction to the control unit 40 if the captured movement data is a right movement, and sends a left direction instruction to the control unit 40 if the movement data is left movement. Send to.

When the left direction instruction is passed from the motion pattern determination unit 47, the control unit 40 passes to the playback unit 46 position information and a playback execution instruction of the music having a higher similarity than the currently played music in the list of search results. At the same time, an instruction to display the metadata (music title, artist name, etc.) of the music is given to the display unit 44.
In addition, when the right direction instruction is passed, the control unit 40 passes the position information and the reproduction execution instruction of the music having a lower similarity than the currently reproduced music in the search result list to the reproduction unit 46 and the position. Information and display instructions are passed to the display unit 44.
If there is no search result, the control unit 40 does not operate even if it receives a left / right direction instruction.
As a result, even when there are a plurality of music pieces in the search result, the music piece can be played even if the similarity degree of the music piece desired by the user is low. Also, similar songs can be played one after another.

  The above-described third to fifth embodiments have been described as single configurations, but the present invention is not limited to this and may be appropriately combined.

It is a block diagram which shows the function structure of the music search apparatus which concerns on Embodiment 1 of this invention. It is a figure which shows the waveform of motion data. It is a figure which shows the waveform of the music data of a music. It is a block diagram which shows the function structure of the music search apparatus which concerns on Embodiment 2 of this invention. It is the figure which imagined the utilization scene of the portable music reproduction terminal by this invention. It is a block diagram which shows the function structure of the portable reproducing | regenerating apparatus which concerns on Embodiment 3 of this invention. 14 is a flowchart for explaining processing until a user plays a desired music piece using the portable playback device according to the third embodiment. It is a block diagram which shows the function structure of the portable reproducing | regenerating apparatus which concerns on Embodiment 4 of this invention. It is a block diagram which shows the function structure of the portable reproducing | regenerating apparatus which concerns on Embodiment 5 of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Music search apparatus, 11 ... Determination part, 12 ... Connector 1, 13 ... Connector 2, 14 ... Connector 3, 15 ... Determination condition setting memory, 16 ... Working memory, 17 ... Feature-value extraction part, 20 ... Sound information storage Unit, 30 ... sensor, 40 ... control unit, 41 ... setting unit, 42 ... mode setting memory, 43 ... motion data input unit, 44 ... display unit, 45 ... input unit, 46 ... reproduction unit, 47 ... motion pattern determination unit .

Claims (22)

  A speech search device comprising: a determination unit that compares motion data of the device with a plurality of speech data and outputs speech data similar to the motion data as a search result.   A feature amount extraction unit that extracts feature amounts from motion data of the device, and compares the feature amount extracted by the feature amount extraction unit with the feature amounts of a plurality of pieces of audio data, and obtains speech data similar to the motion data. A voice search apparatus comprising: a determination unit that outputs a search result.   3. The voice search device according to claim 1, wherein the determination unit calculates a similarity between the motion data and the voice data, and uses the voice data having a similarity higher than a predetermined threshold as a search result. A featured voice search device.   The voice search device according to claim 3, wherein the threshold value can be changed.   A speech search method comprising: comparing motion data of a device with a plurality of speech data, and outputting speech data similar to the motion data as a search result.   A speech search method characterized by comparing feature quantities extracted from motion data of a device and feature quantities of a plurality of speech data, and outputting speech data similar to the motion data as a search result.   In a portable playback device that plays back audio data selected from a plurality of audio data, a motion data input unit that detects and inputs movement of the device, and a sound information storage unit that stores sound information and metadata related to the plurality of audio data A determination unit that compares the motion data input by the motion data input unit with the voice data stored in the sound information storage unit and outputs speech data similar to the motion data as a search result; and the determination unit And a display unit that displays metadata related to the audio data of the search result output in step (b).   In a portable playback device that plays back audio data selected from a plurality of audio data, a motion data input unit that detects and inputs motion of the device, and a feature value that extracts feature values from the motion data input by the motion data input unit An extraction unit; a sound information storage unit that stores sound information and metadata relating to a plurality of audio data; a feature amount extracted by the feature amount extraction unit; and a feature amount of the audio data stored in the sound information storage unit And a determination unit that outputs metadata of audio data similar to the motion data as a search result, and a display unit that displays metadata related to the audio data of the search result output by the determination unit. A portable playback device comprising:   In a portable playback device that plays back audio data selected from a plurality of audio data, a motion data input unit that detects and inputs movement of the device, and a sound information storage unit that stores sound information and metadata related to the plurality of audio data A determination unit that compares the motion data input by the motion data input unit with the voice data stored in the sound information storage unit and outputs speech data similar to the motion data as a search result; and the determination unit And a playback unit that plays back the audio data of the search result output in step (b).   In a portable playback device that plays back audio data selected from a plurality of audio data, a motion data input unit that detects and inputs motion of the device, and a feature value that extracts feature values from the motion data input by the motion data input unit An extraction unit; a sound information storage unit that stores sound information and metadata relating to a plurality of audio data; a feature amount extracted by the feature amount extraction unit; and a feature amount of the audio data stored in the sound information storage unit And a determination unit that outputs audio data similar to the motion data as a search result, and a reproduction unit that reproduces the audio data of the search result output by the determination unit. Portable playback device.   11. The portable playback device according to claim 9, wherein the playback unit plays back audio data having the highest similarity among the search results.   11. The mobile playback device according to claim 9, wherein motion pattern determination is performed to determine whether the motion data input by the motion data input unit is data for search or data for selecting a search result. And when the motion pattern determination unit determines that the search result is selected, the playback unit plays back audio data different from the audio data being played back from the search results. Portable playback device.   13. The portable playback device according to claim 7, wherein when the audio data selected by the playback unit is played back, metadata related to the audio data is displayed.   The portable playback device according to any one of claims 7 to 13, wherein the determination unit calculates a similarity between motion data and audio data, and obtains audio data having a similarity greater than a predetermined threshold as a search result. A portable playback device.   The portable playback device according to claim 14, further comprising a setting unit that sets the threshold, wherein the threshold can be changed.   16. The portable playback apparatus according to claim 7, wherein the sound information storage unit is detachable from the portable playback apparatus.   In a sound reproduction method for reproducing sound data selected from a plurality of sound data, a motion data input step for detecting and inputting a motion of the apparatus, the input motion data, and a plurality of sounds stored in a sound information storage unit A step of comparing the data and outputting speech data similar to the motion data as a search result, a playback step of reproducing the speech data of the search result, and a list of metadata relating to the speech data of the search result And a display step for displaying the sound.   In an audio reproduction method for reproducing audio data selected from a plurality of audio data, a motion data input step for detecting and inputting a motion of the apparatus, a feature amount extracting step for extracting a feature amount from the input motion data, A step of comparing the extracted feature quantity and the feature quantities of a plurality of audio data stored in the sound information storage unit, and outputting audio data similar to the motion data as a search result; and An audio reproduction method comprising: a reproduction step of reproducing audio data; and a display step of displaying a list of metadata related to the audio data of the search result.   5. A music search apparatus, wherein the voice data in the voice search apparatus according to claim 1 is music data.   7. A music searching method according to claim 5, wherein the voice data in the voice searching method according to claim 5 is music data.   The portable playback device according to claim 7, wherein the audio data in the portable playback device is music data.   The music reproducing method according to claim 17 or 18, wherein the audio data in the audio reproducing method is music data.

Images