JP2007264569A - Retrieval device, control method, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To retrieve a model appropriate for a user practicing singing or performance. <P>SOLUTION: When receiving practice voice feature data and a music ID from a karaoke (orchestration without lyrics) device 2, a server device 3 compares the received practice voice feature data with all pieces of model voice feature data stored in a model voice data storage part 32a in association with the music ID and selects a piece of model voice feature data showing the highest similarity to the practice voice feature data from these pieces of model voice feature data and reads out model voice data corresponding to the selected piece of model voice feature data and transmits the read model voice data to the karaoke device 2. The karaoke device 2 reproduces the received model voice data as a model for the user. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a technique for searching for a singing voice or performance sound that serves as a model for a practitioner.

  When a person practicing singing sings along with the accompaniment of a song reproduced by a karaoke device, the singer who sings the song as a song is often imitated. However, a practitioner who is an amateur of singing and a singer who is skilled in singing have a considerable gap in voice quality and singing technology, so the practitioner usually cannot imitate well. Also, if the practitioner's singing skills are too low, it is difficult to improve their singing skills even if they try to imitate and sing. In order to efficiently improve a practitioner's singing skills, an appropriate model that matches the level of the practitioner's voice quality and singing skills is required. This applies not only to the singing of music but also to the performance of musical instruments.

For example, in the technical field for supporting the utterance of the practitioner, the voice of the model (teacher) is converted into the voice of the practitioner (student) and listened to (see Patent Document 1), or the utterance of the practitioner himself. A mechanism has been proposed in which sound is subjected to sound processing such as correcting a spectrum envelope and reproduced (see Patent Document 2).
JP 2002-244547 A JP 2004-133409 A

  The mechanism described in Patent Document 1 requires a procedure in which the practitioner has to select the modeler's voice, which is complicated. Further, in the mechanism described in Patent Document 2, since speech processing is performed on the uttered speech of the practitioner, unnatural pronunciation may occur. The present invention has been made in view of such a background, and an object of the present invention is to provide a technique for searching for an appropriate model for a practitioner with a mechanism different from the conventional one.

  In order to solve the above-described problems, the present invention acquires model sound storage means for storing model sound data representing a typical singing voice or performance sound, and acquisition of practice sound data representing the singing voice or performance sound of a practitioner And a feature of each model sound data stored by the model sound storage unit and a feature of the practice sound data acquired by the acquisition unit, and a feature similar to the feature of the practice sound data There is provided a search device comprising: selection means for selecting exemplary sound data having; and output means for outputting exemplary sound data selected by the selection means.

  The present invention also provides music identification information assigned to a song to be sung or played, model identification information assigned to a model performer singing or performing, and singing voice or performance by each model Model sound storage means that associates and stores model sound data that represents sound, practice sound data that represents the singing voice or performance sound of the practitioner, and practitioner identification information assigned to the practitioner And comparing the characteristics of each model sound data stored in the model sound storage means with the characteristics of the practice sound data acquired by the first acquisition means. First selection means for selecting model sound data having characteristics similar to the characteristics, practitioner identification information acquired by the first acquisition means, and model sound data selected by the first selection means. Identification information storage means for storing in association with the exemplary person identification information stored in the exemplary sound storage means in association with the data, and second acquisition for acquiring the practitioner identification information and the music identification information And identification information stored in the identification information storage means in association with the practitioner identification information acquired by the second acquisition means and associated with the specified exemplary identification information. Second selection means for selecting the model sound data associated with the music identification information acquired by the second acquisition means from among the plurality of model sound data stored in the model sound storage means; And an output means for outputting the model sound data selected by the first selection means or the second selection means.

  The present invention also provides a control method for a search device comprising model sound storage means for storing model singing voice or model sound data representing performance sound, and control means, wherein the control means is a practitioner. A first step of acquiring practice sound data representing the singing voice or performance sound of the sound, a feature of each of the model sound data stored in the model sound storage unit by the control means, and the first step A second step of comparing the characteristics of the acquired practice sound data and selecting model sound data having characteristics similar to the characteristics of the practice sound data; and the control means is selected in the second step. And a third step of outputting model sound data.

The present invention also provides music identification information assigned to a song to be sung or played, model identification information assigned to a model performer singing or performing, and singing voice or performance by each model Model sound storage means for storing model sound data representing sounds in association with each other, identification information storage means for storing the model person identification information and practitioner identification information assigned to the practitioner in association with each other, and control means And a control means for acquiring the practice sound data representing the singing voice or performance sound of the practitioner and the practitioner identification information assigned to the practitioner. And the control means compares the characteristics of each model sound data stored in the model sound storage means with the characteristics of the practice sound data acquired in the first step, A second step of selecting model sound data having characteristics similar to the characteristics of the sound data; and the control means is selected in the second step and the practitioner identification information acquired in the first step. The model sound data selected in the second step is stored in the identification information storage unit in association with the model identification information stored in the model sound storage unit in association with the model sound data. A third step of outputting the learner, a fourth step in which the control means acquires the practitioner identification information and the music piece identification information, and a practitioner identification in which the control means is acquired in the fourth step. The model identification information stored in the identification information storage means in association with the information is identified, and the model sound storage unit is associated with the identified model identification information. A fifth step of selecting model sound data associated with the music identification information acquired by the second acquisition means from among the plurality of model sound data stored in And a sixth step of outputting the model sound data selected in the fifth step.
Furthermore, the present invention may also take the form of a program that causes a computer to realize functions.

  According to the present invention, it is possible to search for a model voice or model performance similar to the singing voice or performance sound of the practitioner, that is, a model suitable for the individual practitioner.

Next, the best mode for carrying out the present invention will be described.
In the following description, a person who practices singing is called a “practicing person”, and a person who performs singing that serves as an example for the practicing person (for example, a singer) is called a “executive person”. Modelers rarely sing according to the content of the score, and in most cases, singing techniques such as intentionally shifting the beginning and end of singing, changing the voice quality and volume, and various singing techniques such as vibrato and fist To express the excitement (feelings) in the song. However, these expressions vary depending on the singer. Therefore, the present embodiment searches for the singing voice of the model person who resembles the singing voice of the practitioner from among the singing voices of the model person, and listens to the practitioner to imitate it. It aims to improve the singing skills.

[1. Constitution]
FIG. 1 is a block diagram showing the overall configuration of a search system 1 according to the present embodiment. The search system 1 includes a plurality of karaoke apparatuses 2a, 2b, and 2c, a server apparatus 3, and a network 4 that connects them. The karaoke devices 2 a, 2 b, 2 c are provided in various households such as ordinary households, karaoke boxes or restaurants, and function as communication devices that perform communication via the network 4. The server device 3 stores the singing voices of a large number of models, and functions as a search device that searches for a singing voice that seems to be appropriate for the practitioner. The network 4 is, for example, an ISDN (Integrated Services Digital Network), the Internet, or an in-store network, and includes a wired section or a wireless section. Although three karaoke apparatuses are illustrated in FIG. 1, the number of karaoke apparatuses included in the search system 1 is not limited to three, and may be more or less. Further, since the karaoke devices 2a, 2b, and 2c all have the same configuration and operation, they are simply collectively referred to as “karaoke device 2” below.

FIG. 2 is a block diagram showing the configuration of the karaoke apparatus 2.
The control unit 21 is, for example, a CPU, and controls each unit of the karaoke apparatus 2 by reading and executing a computer program stored in the storage unit 22. The display unit 23 is a liquid crystal display, for example, and displays various screens such as a menu screen for operating the karaoke device 2 and a karaoke screen in which lyrics telop is superimposed on a background image under the control of the control unit 21. To do. The operation unit 24 includes various keys and outputs a signal corresponding to the pressed key to the control unit 21. The microphone 25 picks up sound produced by the singer. The sound processing unit 26 converts the sound (analog data) collected by the microphone 25 into digital data and outputs the digital data to the control unit 21. The speaker 27 emits sound output from the sound processing unit 26. The communication unit 28 performs data communication with the server device 3 via the network 4 under the control of the control unit 21.

  The storage unit 22 is a large-capacity storage unit such as a hard disk, for example, and includes an accompaniment / lyric data storage area 22a, a practice voice data storage area 22b, and a singing musical score sound data storage area 22c. In the accompaniment / lyric data storage area 22a, accompaniment data in which performance sounds of various musical instruments for accompaniment of music are recorded as the music progresses and lyrics data indicating the lyrics of the music are stored in association with each other. Yes. The accompaniment data has a data format such as MIDI (Musical Instruments Digital Interface) format, and is reproduced when the practitioner sings a karaoke song. The lyrics data is displayed on the display unit 23 as a lyrics telop at the time of the karaoke song. In the practice voice data storage area 22b, voice data A / D converted from the microphone 25 via the voice processing unit 26 is stored as practice voice data. The practice audio data is, for example, in WAVE format or MP3 (MPEG Audio Layer-3) format. The musical score data storage area 22c stores musical score data (for example, MIDI format) representing the pitch and timing of the singing portion defined by the musical score of the music. This music score data is used to extract various techniques such as “vibrato”, “shakuri”, “fist”, “farset”, “tsukkomi”, “for”, “breathing”, etc. Used.

Next, FIG. 3 is a block diagram showing a configuration of the server device 3.
In FIG. 3, the control unit 31 is, for example, a CPU, and controls each unit of the server device 3 by reading and executing a computer program stored in the storage unit 32. The storage unit 32 is a large-capacity storage unit such as a hard disk. The storage unit 32 includes a model voice data storage area 32a and a practice voice feature data storage area 32b. The communication unit 33 performs data communication with the karaoke apparatus 2 through the network 4 under the control of the control unit 31.

Here, the contents stored in the storage unit 32 will be described in detail.
FIG. 4 is a diagram illustrating an example of data stored in the model audio data storage area 32a. As shown in FIG. 4, in the model voice data storage area 32a, a song ID (identification: identification information) assigned to the song, a modeler ID assigned to the modeler, and the modeler sang the song. The model voice data representing the singing voice at the time and the model voice feature data representing the characteristics of the singing voice of the model person are stored in association with each other. The song ID is identification information such as a song name or karaoke song number. The model ID is, for example, the model name (singer name) of the model. The model voice data is recorded in advance, is transmitted from the server apparatus 3 to the karaoke apparatus 2, and is reproduced by the karaoke apparatus 2 as a model voice for the practitioner. The model voice feature data includes the pitch (pitch) of the model person's singing voice, the sounding timing, the spectrum, the power (volume), the type of technique used for the singing, and its section. In the example of FIG. 4, for example, the model voice data representing the voice when the model person “XXX” sings the song with the song ID “g01” and the model voice feature data representing the feature are associated with each other. The case where it is stored is illustrated.

  In the practice voice feature data storage area 32b of the storage unit 32, practice voice feature data representing features of the practice voice data transmitted from the karaoke apparatus 2 is stored. The practice voice feature data includes the pitch, pronunciation timing, spectrum, power, and type of technique used for singing and its section, as in the above-described exemplary voice feature data.

[2. Operation]
Next, the operation of this embodiment will be described.
In the sequence diagram of FIG. 5, the practitioner operates the operation unit 24 of the karaoke device 2 to select a song ID of a song that the user wants to sing, and instructs playback of the karaoke accompaniment. In response to this operation, the control unit 21 starts karaoke accompaniment (step S1). That is, the control unit 21 reads the accompaniment data corresponding to the designated music ID from the accompaniment / lyric data storage area 22a and supplies the accompaniment data to the audio processing unit 26. The audio processing unit 26 converts the accompaniment data into an analog signal. Then, sound is emitted from the speaker 27. At the same time, the control unit 21 displays a message prompting singing such as “Please sing along with the accompaniment” on the display unit 23, then reads out the lyric data from the accompaniment / lyric data storage area 22a and reads the lyrics telop. Is displayed on the display unit 23. The practitioner sings along with the accompaniment emitted from the speaker 27 while referring to the displayed lyrics telop. At this time, the practitioner's voice is picked up by the microphone 25, converted into a voice signal, and output to the voice processing unit 26. The practice voice data A / D converted by the voice processing unit 26 is stored (recorded) in the practice voice data storage area 22b of the storage unit 22 together with information indicating the elapsed time from the start of accompaniment (step S2). ).

  When the reproduction of the accompaniment data ends, the control unit 21 ends the process of recording the practitioner's singing voice. And the control part 21 isolate | separates the practice sound data memorize | stored in the practice sound data storage area 22b into the frame unit of predetermined time length, and calculates a pitch, a spectrum, and power for each frame unit (step S3). For example, FFT (Fast Fourier Transform) may be used to calculate the spectrum.

  Next, the control unit 21 extracts a technique from the practice voice data (step S4). As described above, the techniques include “vibrato”, “shakuri”, “fist”, “farset”, “push”, “for”, and “breathing”. Among these, “vibrato” is a technique that produces a timbre-like tone by raising and lowering the pitch of the sound only slightly. “Shikkuri” is a technique in which sound is generated from a sound lower than the target sound, and the pitch is smoothly brought close to the target sound. “Fist” is a technique of performing a undulating curl that is decoratively added. “Falset” is a technique of singing with a so-called “back voice”. “Tsukumi” is a technique in which singing is performed at a timing earlier than the original timing. “Tame” is a technique in which singing is made later than the original timing. “Respiration” means the timing when the practitioner takes a breath.

  First, the control unit 21 specifies (detects) a section in which each technique is used in the practice voice data. For example, “vibrato” and “shrimp” can be detected based on the pitch of the model voice data. Further, “fist” and “falset” can be detected based on the spectrum of the model voice data. Further, “for” and “tsukkomi” can be detected based on the pitch of the model voice data and the musical score sound data stored in the singing musical score sound data storage area 22c. Further, “breathing” can be detected based on the power of the model voice data and the musical score data stored in the singing musical score data storage area 22c.

A specific detection method is as follows.
Based on the correspondence between the practice voice data and the score sound data and the pitch calculated from the practice voice data, the control unit 21 starts the sound included in the practice voice data and the score sound data corresponding to the sound. The section where the start time of the sound is different is specified. Here, the control unit 21 corresponds to a section in which the pitch change timing of the practice voice data appears earlier than the pitch change timing of the musical score sound data, that is, the start time of the sound included in the practice voice data. For a section earlier than the start time of the sound of the musical score data, this section is specified as a section in which the “Tsukumi” singing technique is used. The control unit 21 associates the section information of the section specified in this way with identification information indicating “push”.

  Conversely, the control unit 21 determines that the pitch change timing of the practice voice data is a change in the pitch of the score sound data based on the correspondence between the practice voice data and the score sound data and the pitch calculated from the practice voice data. A section that appears later than the timing, that is, a section in which the start time of the sound included in the practice sound data is later than the start time of the sound of the musical score sound data corresponding to the sound, and the detected section is It is specified that it is a section in which the singing technique is used.

  Further, the control unit 21 analyzes a pattern of temporal change of the pitch calculated from the practice voice data, and detects a section where the pitch continuously fluctuates within a predetermined range above and below the central frequency. The detected section is identified as a section in which the “vibrato” singing technique is used.

  Further, the control unit 21 analyzes a pattern of temporal change of the pitch calculated from the practice voice data, detects a section where the pitch continuously changes from a low pitch to a high pitch, ”Is identified as the section in which the singing technique is used. This process may be performed based on the correspondence with the musical score data. That is, the control unit 21 may detect a section in which the pitch of the practice voice data continuously approaches the pitch of the score sound data from a low pitch based on the correspondence relationship between the practice voice data and the score sound data. .

  Further, the control unit 21 is a section in which the musical score sound data is sound based on the correspondence between the practice voice data and the musical score sound data and the power calculated from the practice voice data. A section whose value is smaller than a predetermined threshold is detected, and the detected part is specified as a section of “breathing”.

  In addition, the control unit 21 analyzes the temporal change pattern of the spectrum calculated from the practice voice data, and detects and detects a section where the spectral characteristics are rapidly transitioning to the predetermined change state. The section is identified as a section in which the “Falset” singing technique is used. Here, the predetermined change state is a state in which the harmonic component of the spectrum characteristic is extremely reduced. For example, in the case of a local voice, many harmonic components are included, but when a false set is used, the magnitude of the harmonic components becomes extremely small. In this case, the control unit 21 may also refer to whether or not the pitch has changed significantly upward. The falset is sometimes used even when generating the same pitch as the local voice, but is generally a technique used when generating high-pitched sounds that cannot be generated by the local voice. Therefore, the “false set” may be detected only when the pitch of the practice voice data is equal to or higher than a predetermined pitch. In addition, since the pitch range using falsets is generally different between male voices and female voices, gender detection is performed based on the range of the practice voice data and formants detected from the practice voice data. An area may be set.

  In addition, the control unit 21 detects a section in which the mode of change of the spectrum characteristic is variously switched in a short time, and identifies the detected part as a part where the “fist” singing technique is used. In the case of “fist”, it is a singing technique that adds a taste that can be changed by changing the voice color and utterance method in a short section, so the spectral characteristics change variously in the section where this technique is used. .

  As described above, the control unit 21 detects the section in which the technique is used from the practice voice data, and associates the section information indicating the detected section with the type information indicating the singing technique. Then, the control unit 21 generates practice voice feature data including the pitch, spectrum, and power calculated in step S3 and the section information and type information generated in step S4 (step S5). At this time, the control unit 21 also calculates the sound generation timing from the pitch and includes this in the practice voice feature data. And the control part 21 transmits the produced practice audio | voice feature data to the server apparatus 3 from the communication part 28 with music ID (step S6).

  Upon receiving the practice voice feature data and the song ID, the control unit 31 of the server device 3 receives all the practice voice feature data received and all the model persons stored in association with the song ID in the model voice data storage area 32a. The voice feature data is compared, and the model voice feature data having the highest similarity with the practice voice feature data is selected (step S7). More specifically, the control unit 31 integrates the difference between the pitch represented by the practice voice feature data and the pitch represented by each model voice feature data over the entire area from the start to the end of the singing. Similarly, the control unit 31 integrates the difference between the power represented by the practice voice feature data and the power represented by each model voice feature data over the entire area from the start to the end of the singing. The same applies to the spectrum and timing. In addition, the control unit 31 also uses the section represented by the section information of each technique represented by the practice voice feature data and the section represented by the section information of each technique represented by each model voice feature data in the same manner as described above. Integrate the difference between and. Then, the control unit 31 accumulates the integrated value obtained as described above for each model voice feature data, and sets the model voice feature data having the smallest accumulated value as the similarity to the practice voice data. Choose as the highest.

  Next, the control unit 31 reads out model voice data corresponding to the selected model person voice feature data from the model voice data storage area 32a (step S8), and data that can be reproduced by the karaoke apparatus 2 by the read out model voice data. It is transmitted (output) in the form (step S9). The control unit 21 of the karaoke apparatus 2 reproduces the received model voice data (step S10). That is, the control unit 21 supplies the model voice data to the voice processing unit 26, and the voice processing unit 26 converts the model voice data into an analog signal and emits sound from the speaker 27. At this time, the control unit 21 causes the display unit 23 to display a message “The singing teacher that suits you is Mr. XXX. Let's listen and imitate.” Thereby, the practitioner can hear the singing voice of the model suitable for his / her singing, and can improve his / her singing by using it as a model.

  According to the embodiment described above, the singing voice of the model person similar to the singing voice of the practitioner can be searched and provided to the practitioner as a model.

[3. Modified example]
The above-described embodiment may be modified as follows.
[3-1] In the above-described embodiment, the example of searching for the singing voice (exemplary voice) of the model person who matches the singing voice (practice voice) of the practitioner has been described as an example. The performance sound (exemplary performance sound) of the model person that matches the performance sound (practice performance sound) of the practitioner's instrument may be searched. In this case, model performance sound data representing the performance sound of the model person is used instead of the singing voice of the model person described above, and practice performance sound data representing the performance sound of the practice person is used instead of the practice sound data. The accompaniment / lyric data storage area 22a stores performance data of an instrument (for example, bass or drum) other than the instrument (for example, guitar) to be practiced, and the singing score sound data storage area 22c performs performance on a score. Musical score data specified as a sound is stored. Based on these data, the control unit 31 of the server device 3 searches for model performance sound data similar to the practice performance sound data through the same processing as described above.
Thus, this invention memorize | stores the model sound data showing the singing voice or performance sound used as a model, and when the practice sound data showing the practitioner's singing voice or performance sound is acquired, each memorize | stored The feature of the model sound data is compared with the feature of the acquired practice sound data, and the model sound data having a feature similar to the feature of the practice sound data is selected and output.

[3-2] In the above-described embodiment, after the practitioner sings the desired music, the singing voice of the model person similar to the singing voice is searched. Therefore, the practitioner cannot obtain a singing voice that serves as an example for the song unless the song is sung at least once. When the practitioner feels such procedures complicated, the following modifications may be made. This modification focuses on the fact that if the singing voice of the practitioner and the singing voice of the model person are similar, the similarity between them is high even if the music is different.
As shown in FIG. 6, the storage unit 32 of the server device 3 has a corresponding ID storage area 32 c in addition to the above-described exemplary voice data storage area 32 a and practice voice feature data storage area 32 b. In this correspondence ID storage area 32c, as shown in FIG. 7, the practitioner ID assigned to the practitioner and the modeler ID assigned to the modeler are stored in association with each other. When the practitioner sings a song (for example, song ID “g01”) in the search system 1 and searches for model voice data that serves as a model, the model ID of the model voice data searched at that time is The corresponding ID is stored in the corresponding ID storage area 32c in association with the practitioner's ID. Thereafter, when the practitioner tries to search for model voice data for another music piece (for example, the music piece ID “g02”), it is stored in the corresponding ID storage area 32c in association with the practitioner ID of the practitioner. The model ID associated with the music ID “g02” is identified from among the plurality of model voice data stored in the model voice data storage area 32a in association with the identified model ID. Audio data is searched.

A specific operation example is shown in FIG. In FIG. 8, the same operations as those in FIG. 5 are denoted by the same reference numerals.
Before step S1 in FIG. 8, the practitioner inputs his / her practitioner ID to the karaoke apparatus 2 in addition to the music ID. And in step S6 ', the control part 21 of the karaoke apparatus 2 transmits practice person ID to the server apparatus 3 from the communication part 28 with practice audio | voice characteristic data and music ID. Upon receiving the practice voice feature data, the song ID, and the trainer ID, the control unit 31 of the server device 3 selects the model voice feature data having the highest similarity to the practice voice feature data in step S7, and then step S11. As an example, the trainer ID received from the karaoke apparatus 2 is stored in the corresponding ID storage area 32c in association with the model ID associated with the selected model voice feature data. Then, the control unit 31 reads out the model voice data corresponding to the selected model person voice feature data from the model voice data storage area 32a (step S8), and transmits (outputs) the read out model voice data to the karaoke apparatus 2. (Step S9). The control unit 21 of the karaoke apparatus 2 reproduces the received model voice data (step S10).

  Thereafter, when the practitioner inputs his / her practitioner ID together with another music ID to the karaoke apparatus 2, the karaoke apparatus 2 accepts this input (step S12). Then, the control unit 21 transmits the input music ID and practitioner ID from the communication unit 28 to the server device 3. The control part 31 of the server apparatus 3 will identify model person ID matched with the practicer ID in the corresponding ID storage area 32c, if music ID and practitioner ID are received (step S13). And the control part 31 reads the model audio | voice data corresponding to the said music ID and model person ID from the model audio | voice data storage area 32a (step S14), and transmits the read model audio | voice data to the karaoke apparatus 2. FIG. The control unit 21 of the karaoke apparatus 2 reproduces the received model voice data (step S15). In this way, the practitioner can search for a singing voice that serves as an example for the music without having to bother singing the music to be practiced.

  In addition, in order for the search system 1 according to the above-described embodiment to function effectively, a sufficient number of exemplary voice data must be prepared for all musical pieces. This is because if there is little model voice data, model voice data that matches the singing voice of the practitioner cannot be obtained. Therefore, in the process in which the search system 1 operates, the practice voice data itself stored by the karaoke device 2 may be stored in the server device 3 as model voice data representing the model person's singing voice.

[3-3] In the embodiment described above, one model voice feature data having the highest similarity to the practice voice feature data is selected, but the number of model voice feature data to be selected is limited to one. Instead, a plurality of model voice feature data may be selected in descending order of similarity, and model voice data corresponding to each selected model voice feature data may be output. For example, the control unit 31 transmits (outputs) the modeler ID (singer name) assigned to the selected plurality of model voice data to the karaoke device 2 and outputs these modeler IDs to the karaoke device 2 in a list format. Display. When the practitioner designates a desired model ID (singer name) from this list, the control unit 21 of the karaoke apparatus 2 transmits the model ID to the server apparatus 3. When receiving the model ID from the karaoke apparatus 2, the control unit 31 transmits (outputs) the model voice data of the model person assigned with the model ID to the karaoke apparatus 2.

[3-4] In the embodiment, the similarity is determined using the integrated value of the difference between the practice voice feature data and each model voice feature data. For example, in the multidimensional space, the practice voice feature data The Euclidean distance between the coordinates of each of the model voice feature data may be calculated, and the model voice feature data having the smallest Euclidean distance may be selected as having the highest similarity.

[3-5] In the above-described embodiment, all of the pitch, timing, power, spectrum, and technique of the voice are used as the model voice feature data and the practice voice feature data, but it is also possible to use at least one of these. Further, other characteristic elements may be used. Also, it may be possible for the practitioner to select which of these feature elements to use using the operation unit 24. Similarly, the practitioner may select any one of various techniques.

[3-6] In the embodiment described above, the practice voice feature data is generated by the control unit 21 of the karaoke apparatus 2, but instead of this, the control unit 31 of the server apparatus 3. May also be generated. Alternatively, the control unit 21 of the karaoke apparatus 2 may prompt the practice voice feature data to be input, and the practice voice feature data prepared in advance by the practitioner may be inputted. In this case, for example, the control unit 21 causes the display unit 23 to display a screen that prompts the user to input practice voice feature data, and the practitioner sends the practice voice to the karaoke apparatus 2 via an interface such as a USB (Universal Serial Bus). What is necessary is just to input feature data. In this case, the practice voice feature data may be generated in advance by a device such as a personal computer. At this time, as in the above-described embodiment, the personal computer collects the voice of the practitioner with the microphone, analyzes the collected voice, and generates practice voice feature data. Further, an RFID reader may be provided in the karaoke device 2 so that the RFID in which the practice voice feature data is written is read by the RFID reader of the karaoke device 2.

[3-7] The form in which the server apparatus 3 outputs the model voice data is not limited to the transmission to the karaoke apparatus 2, but the form in which the model voice data is attached to the e-mail addressed to the practitioner's mail terminal. There may be. Also, the model voice data may be output and stored on a storage medium. In this case, the practitioner listens to the model voice data by reading it from the storage medium and playing it using a computer. be able to. Further, when the model voice data is reproduced by the karaoke apparatus 2, it is not necessary to reproduce the first to the last of the music, and only a part of the music may be reproduced. For example, if only a singing portion having a low feature similarity is reproduced, the practitioner can recognize which singing portion should be focused on.

[3-8] In the embodiment, the server device 3 extracts the model voice feature data from the model voice data and stores the model voice data in advance. Instead, the server apparatus 3 stores only the model voice data. In addition, the model voice feature data may be extracted from the model voice data every time it is necessary to perform a search. The model voice data and the practice voice data are data in the WAVE format or the MP3 format. However, the data format is not limited to this, and any format may be used as long as the data indicates voice. Also good.

[3-9] In the above-described embodiment, the search system 1 in which the karaoke device 2 and the server device 3 are connected via a communication network realizes all the functions according to the present embodiment. On the other hand, three or more devices connected via a communication network may share the above functions, and a system including the plurality of devices may realize the system of the embodiment. Alternatively, one device may realize all of the above functions.

[3-10] A program executed by the control unit 21 of the karaoke device 2 or the control unit 31 of the server device 3 in the above-described embodiment is a magnetic tape, a magnetic disk, a flexible disk, an optical recording medium, a magneto-optical recording medium, It can be provided in a state where it is stored in a recording medium such as a CD (Compact Disk) -ROM, a DVD (Digital Versatile Disk), or a RAM. It is also possible to download to the karaoke apparatus 2 or the server apparatus 3 via a network such as the Internet.

1 is a block diagram showing an overall configuration of a search system according to an embodiment of the present invention. It is a block diagram which shows the structure of a karaoke apparatus. It is a block diagram which shows the structure of a server apparatus. It is a figure which shows an example of the data memorize | stored by the server apparatus. It is a sequence diagram which shows operation | movement of embodiment. It is a block diagram which shows the structure of the server apparatus concerning a modification. It is a figure which shows an example of the data memorize | stored by the server apparatus. It is a sequence diagram which shows operation | movement of the modification.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Search system, 2a, 2b, 2c ... Karaoke apparatus, 3 ... Server apparatus, 4 ... Network, 21 ... Control part, 22 ... Memory | storage part, 23 ... Display part, 24 ... Operation part, 25 ... Microphone, 26 ... Audio processing unit 27 ... speaker 28 ... communication unit 31 ... control unit 32 ... storage unit 33 ... communication unit

Claims (10)

An exemplary sound storage means for storing exemplary sound data representing an exemplary singing voice or performance sound;
Acquisition means for acquiring practice sound data representing the singing voice or performance sound of the practitioner;
The model sound having the characteristics similar to the characteristics of the practice sound data by comparing the characteristics of each model sound data stored by the model sound storage means with the characteristics of the practice sound data acquired by the acquisition means. A selection means for selecting data;
An output unit that outputs the model sound data selected by the selection unit. Music identification information assigned to a song to be sung or played, model identification information assigned to a model person who performs singing or playing as a model, and model sound data representing singing voice or performance sound by each model person A model sound storage means for storing
First acquisition means for acquiring practice sound data representing the singing voice or performance sound of the practitioner and practitioner identification information assigned to the practitioner;
The characteristics of each model sound data stored by the model sound storage means are compared with the characteristics of the practice sound data acquired by the first acquisition means, and the characteristics similar to the characteristics of the practice sound data are compared. First selection means for selecting exemplary sound data having;
The practitioner identification information acquired by the first acquisition unit and the modeler identification information stored in the model sound storage unit in association with the model sound data selected by the first selection unit. Identification information storage means for storing in association with each other;
Second acquisition means for acquiring the practitioner identification information and the music identification information;
The model identification information stored in the identification information storage unit in association with the practitioner identification information acquired by the second acquisition unit is specified, and the model identification information is associated with the specified model identification information. Second selection means for selecting model sound data associated with the music identification information acquired by the second acquisition means from among the plurality of model sound data stored in the sound storage means;
A search apparatus comprising: output means for outputting model sound data selected by the first selection means or the second selection means. A communication unit that performs communication via a network with a communication device including a sound collection unit and a reproduction unit;
The acquisition means receives the practice sound data collected by the sound collection means of the communication device by the communication means,
The search device according to claim 1, wherein the output unit transmits the exemplary sound data in a data format reproducible by the reproduction unit of the communication device from the communication unit to the communication device. The selection means selects a plurality of model sound data from the model sound data stored in the model sound storage means in descending order of similarity to the characteristics of the practice sound data acquired by the acquisition means. ,
The output means first outputs the model person identification information assigned to the plurality of model sound data selected by the selection means, and then the model person identification information designated among these model person identification information is The assigned model sound data is output. The search device according to claim 1, wherein: The first selection unit includes a plurality of model sound data stored in the model sound storage unit, in order of decreasing similarity to the characteristics of the practice sound data acquired by the first acquisition unit. Select model sound data,
The output means first outputs the model person identification information assigned to the plurality of model sound data selected by the first selection means, and then the model person specified in the model person identification information. Output the model sound data to which the identification information is assigned,
The identification information storage means associates the practitioner identification information acquired by the first acquisition means with the exemplary person identification information specified among the plurality of exemplary person identification information output by the output means. The search device according to claim 2, wherein the search device is stored.   The feature of the singing voice or performance sound is at least one of pitch, timing, spectrum, power, and technique data indicating the type and section of the technique used for singing or performance. The search device according to claim 1 or 2. An exemplary sound storage means for storing exemplary sound data representing an exemplary singing voice or performance sound, and a control method for a search device comprising a control means,
A first step in which the control means obtains practice sound data representing a practicing song voice or performance sound;
The control means compares the characteristics of each model sound data stored in the model sound storage means with the characteristics of the practice sound data acquired in the first step, and determines the characteristics of the practice sound data. A second step of selecting exemplary sound data having similar characteristics;
The control means comprises a third step of outputting the model sound data selected in the second step. Music identification information assigned to a song to be sung or played, model identification information assigned to a model person who performs singing or playing as a model, and model sound data representing singing voice or performance sound by each model person A search device comprising: model sound storage means for storing the reference information, identification information storage means for storing the reference person identification information and the practitioner identification information assigned to the practitioner, and control means Control method,
A first step in which the control means acquires practice sound data representing the singing voice or performance sound of the practitioner and practitioner identification information assigned to the practitioner;
The control means compares the characteristics of each model sound data stored in the model sound storage means with the characteristics of the practice sound data acquired in the first step, and determines the characteristics of the practice sound data. A second step of selecting exemplary sound data having similar characteristics;
The control means is associated with the trainer identification information acquired in the first step and the model sound data selected in the second step, and is stored in the model sound storage means. A third step of outputting the exemplary sound data selected in the second step while storing the identification information in the identification information storage unit in association with the information;
A fourth step in which the control means acquires the practitioner identification information and the music piece identification information;
The control means specifies model identification information stored in the identification information storage means in association with the practitioner identification information acquired in the fourth step, and associates with the specified model identification information. A fifth step of selecting the model sound data associated with the music identification information acquired by the second acquisition means from among the plurality of model sound data stored in the model sound storage means; ,
The control method comprises: a sixth step of outputting the model sound data selected in the fifth step. In a computer equipped with model sound storage means for storing model sound data representing an exemplary singing voice or performance sound,
An acquisition function for acquiring practice sound data representing a practicing song voice or performance sound;
A model sound having characteristics similar to the characteristics of the practice sound data by comparing the characteristics of each model sound data stored by the model sound storage means with the characteristics of the practice sound data acquired by the acquisition function A selection function to select data,
An output function for outputting model sound data selected by the selection function. Music identification information assigned to a song to be sung or played, model identification information assigned to a model person who performs singing or playing as a model, and model sound data representing singing voice or performance sound by each model person A computer comprising: model sound storage means for storing the information in association with each other; and identification information storage means for storing the model identification information and the practitioner identification information assigned to the practitioner in association with each other.
A first acquisition function for acquiring practice sound data representing the singing voice or performance sound of the practitioner and practitioner identification information assigned to the practitioner;
The feature of each model sound data stored by the model sound storage means is compared with the feature of the practice sound data acquired by the first acquisition function, and the feature is similar to the feature of the practice sound data. A first selection function for selecting model sound data having;
Trainer identification information acquired by the first acquisition function and model identification information stored in the model sound storage means in association with the model sound data selected by the first selection function. A writing function to be associated and stored in the identification information storage means;
A second acquisition function for acquiring the practitioner identification information and the music identification information;
The model identification information stored in the identification information storage means in association with the practitioner identification information acquired by the second acquisition function is specified, and the model identification information is associated with the specified model identification information. A second selection function for selecting model sound data associated with the music identification information acquired by the second acquisition means from among the plurality of model sound data stored in the sound storage means;
A program for realizing an output function for outputting model sound data selected by the first selection function or the second selection function.

Images