JP2010060838A - Musical piece quality determination device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To compare qualities of a plurality of musical contents. <P>SOLUTION: A car audio device includes: a tuner part which receives a broadcast radiowave; a musical content extraction part 21 which extracts a musical content from the received broadcast wave; an HDD 10 where the extracted musical content is stored; and a comparison part 41 which, if a musical content identical to the extracted musical content is stored in the HDD 10, compares a quality level of the extracted musical content with that of the stored musical content, wherein the comparison part 41 includes an evaluation index determination part 43 for determining a plurality of evaluation indexes per musical content and a quality level calculation part 45 for calculating quality levels resulting from weighting the plurality of evaluation indexes. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a music quality determination apparatus, a music quality determination method, and a music quality determination program, and more particularly to a music quality determination apparatus, a music quality determination method, and a music quality determination program for determining the quality of music content included in a broadcast signal.

  In recent years, a technique for automatically discriminating music from audio broadcast in a radio broadcast program is known. However, since the music is automatically recorded, stations with the same title are recorded twice. On the other hand, the quality of music broadcast on a radio broadcast program varies depending on the timing of receiving the radio broadcast. For example, when only a part of the music is broadcast, the quality of the music differs depending on the presence or absence of a comment generated by a disc jockey, the difference in noise level due to the difference in radio wave state at the time of reception, and the like.

  For example, in Japanese Patent Application Laid-Open No. 2005-249991 (Patent Document 1), even when a plurality of music pieces having the same content are stored in a state where the quality is different or the time axis is shifted, A technique for detecting music with the same content is described.

However, the technique described in Japanese Patent Application Laid-Open No. 2005-249991 can detect music pieces having the same content, but cannot determine which quality of a plurality of music pieces is high.
Japanese Patent Laid-Open No. 2005-249991

  The present invention has been made to solve the above-described problems, and one of the objects of the present invention is to provide a music quality determination apparatus capable of comparing the quality of a plurality of music contents.

  According to an aspect of the present invention, in order to achieve the above-described object, a music quality determination device includes: a receiving unit that receives broadcast radio waves; an extraction unit that extracts music content from the received broadcast radio waves; A comparison of comparing the quality level of the extracted music content with the quality level of the stored music content when the storage means storing the music content and the same music content as the extracted music content are stored in the storage means And a comparison unit includes an evaluation index determination unit that determines a plurality of evaluation indexes for each piece of music content, and a calculation unit that calculates a quality level weighted to the plurality of evaluation indexes.

  According to this aspect, a plurality of evaluation indexes are determined for each music content, a quality level weighted to the determined plurality of evaluation indexes is calculated, and the quality levels of the same two music contents are compared. For this reason, since the quality levels of two music contents are compared with a plurality of evaluation indexes, the quality of the two music contents can be compared. As a result, it is possible to provide a music quality determination device capable of comparing the quality of a plurality of music contents.

  Preferably, the calculation means further includes a setting means for setting a plurality of weighting coefficients corresponding to each of the plurality of evaluation indexes, and when a change by the user is accepted, a value obtained by changing the plurality of weighting coefficients by the user. Set to.

  According to this aspect, since the weighting coefficient can be set, the music content having the quality according to the user's preference can be selected.

  According to another aspect of the present invention, the music quality determination device includes: a receiving unit that receives broadcast radio waves; an extraction unit that extracts music content from the received broadcast radio waves; and an audio signal level or Cut point detection means for detecting a time point when the change amount is equal to or greater than a predetermined value as a cut point; and determination means for determining a noise level of the extracted music content based on the number of detected cut points. .

  Preferably, the evaluation index determination unit includes a noise likelihood calculation unit that calculates a noise likelihood determined by a relationship with teacher data of noise determined in advance using a likelihood function for the music content, and a calculated noise Noise amount evaluation index determining means for determining a noise amount evaluation index based on the maximum likelihood value.

  Preferably, the evaluation index determination means includes a noise likelihood calculation means for calculating a noise likelihood determined by a relationship with teacher data of noise determined in advance using a likelihood function for the music content, and a calculated noise Noise ratio evaluation index determination means for determining a noise ratio evaluation index based on the average value of likelihood.

  Preferably, the evaluation index determining means calculates the music likelihood calculating means for calculating the music likelihood determined in relation to the teacher data of the music predetermined using the likelihood function for the music content, and the calculated music Music amount evaluation index determination means for determining a music amount evaluation index based on the maximum likelihood value.

  Preferably, the evaluation index determining means calculates the music likelihood calculating means for calculating the music likelihood determined in relation to the teacher data of the music predetermined using the likelihood function for the music content, and the calculated music Music ratio evaluation index determination means for determining a music ratio evaluation index based on the average value of likelihood.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the following description, the same parts are denoted by the same reference numerals. Their names and functions are also the same. Therefore, detailed description thereof will not be repeated.

  FIG. 1 is a diagram illustrating an example of a basic configuration of a car audio device according to the present embodiment. Referring to FIG. 1, a car audio device is a central processing unit (CPU) 7 that controls the entire car audio device, a memory 8, a tuner unit 1 that receives broadcast radio waves, and converts an analog signal into a digital signal. A / D converter 2 that compresses, a digital audio signal MP3 (MPEG Audio Layer-3) codec 3, a D / A converter 4 that converts a digital signal into an analog signal, a speaker 5, and a large capacity A hard disk drive (HDD) 10 as a storage device, an HDD interface unit 9, an external storage device 11, and an operation unit 12 are included.

  The tuner unit 1 is controlled by the CPU 7 and receives broadcast radio waves such as television broadcasts or radio broadcasts, receives radio waves of a broadcast station selected from the received broadcast radio waves, and receives the received broadcast radio waves as an audio signal. Demodulate. The tuner unit 1 outputs the demodulated audio signal to the A / D converter 2 and outputs the radio wave intensity of the received broadcast radio wave to the CPU 7.

  The A / D converter 2 converts the audio signal input from the tuner unit 1 into digital audio data, and outputs the audio data to the CPU 7 and the MP3 codec 3. The HDD interface unit 9 is an interface for connecting the CPU 7 to the HDD 10.

  The MP3 codec 3 is controlled by the CPU 7 and compresses the digital audio data input from the A / D converter 2 according to an instruction from the CPU 7 to generate encoded data. The MP3 codec 3 stores the generated encoded data in the HDD 10 via the HDD interface unit 9. The MP3 codec 3 is controlled by the CPU 7 and reads the encoded data stored in the HDD 10 in accordance with an instruction from the CPU 7, decodes the read encoded data, and generates audio data. The MP3 codec 3 outputs the generated audio data to the D / A converter 4.

  The D / A converter 4 converts the digital audio data input from the MP3 codec 3 into an analog audio signal. Then, the audio signal converted into analog is output to the speaker 5. Note that the speaker 5 receives an audio signal amplified by an amplifier (not shown).

  The memory 8 includes a ROM (Read Only Memory) for storing a program executed by the CPU 7 and a RAM (Random Access Memory) used as a work area of the CPU 7.

  The external storage device 11 is mounted with a CD-ROM (Compact Disc-Read Only Memory) 11A, and is an interface for the CPU 7 to access data stored in the CD-ROM 11A. The CPU 111 reads the program stored in the CD-ROM 11A mounted on the external storage device 11, stores the read program in the RAM, and executes it. The recording medium for storing the program to be executed by the CPU 111 is not limited to the CD-ROM 11A, but is a flexible disk, cassette tape, optical disk (MO (Magnetic Optical Disc) / MD (Mini Disc) / DVD (Digital Versatile Disc). )), A medium such as an IC card, an optical card, a mask ROM, an EPROM (Erasable Programmable ROM), an EEPROM (Electronically EPROM), or the like. The program here includes not only a program that can be directly executed by the CPU 7, but also a source program, a compressed program, an encrypted program, and the like.

  The operation unit 12 includes a plurality of keys and accepts an operation input by the user. The operation unit 12 may include a liquid crystal display device and a touch panel made of a transparent member superimposed thereon.

  FIG. 2 is a functional block diagram illustrating an example of an outline of the functions of the CPU. Referring to FIG. 2, the CPU 7 included in the car audio device includes a music content extraction unit 21 that extracts a music part (music content) from audio data, a cut point detection unit 23 that detects a cut point, and a noise likelihood. A noise likelihood calculating unit 25 for calculating, a music likelihood calculating unit 27 for calculating music likelihood, a radio wave intensity acquiring unit 29 for acquiring the radio wave intensity of a broadcast signal including music content, and a program for acquiring program information The information acquisition part 31, the music length acquisition part 33 which acquires the length of a music, the comparison part 41, and the setting reception part 35 which receives the setting by a user are included.

  The music content extraction unit 21 extracts a music part including music from the audio data input from the A / D converter 2 as music content. The technology for extracting music content from audio data, for example, detects a time point when the change amount of the audio signal level is a predetermined value or more as a cut point, calculates a feature amount in the frequency domain of the audio signal only in the vicinity of the cut point, A technique for determining the attribute of the cut point based on the calculated frequency feature amount and detecting the start point and the end point of the music section based on the interval between the attribute and the sampling point can be used. The music content extraction unit 21 outputs the extracted music content to the cut point detection unit 23, the noise likelihood calculation unit 25, and the music likelihood calculation unit 27.

  The cut point detection unit 23 detects a time point when the amount of change in the audio signal level of the music content is equal to or greater than a predetermined value, and outputs the number of cut points extracted from the music content to the comparison unit 41.

  The noise likelihood calculating unit 25 calculates MFCC (Merkel frequency cepstrum count) as a frequency feature amount of the music content in a predetermined period, and calculates the likelihood of the calculated feature amount and the teacher data of the reference noise as noise. The likelihood is calculated and the noise likelihood is output to the comparison unit 41. For example, white noise MFCC included in FM radio broadcasting may be used as noise teacher data. Since the noise likelihood calculation unit 25 calculates and outputs one noise likelihood in a predetermined period, the noise likelihood calculation unit 25 outputs the noise likelihood at predetermined period intervals.

  In addition, although the example which calculates noise likelihood is shown here, you may make it compare the frequency component of the same music content. In the same section of the same music content, the frequency component due to the music portion should be the same, and the difference generated when compared can be regarded as the frequency component due to the noise portion. In other words, when the frequency components of the same music content are compared, the more frequency components are included, the more noise is included, and the noise level can be regarded as high.

  The music likelihood calculating unit 27 calculates MFCC (Merkel frequency cepstrum count) as the frequency feature amount of the music content in a predetermined period, and calculates the likelihood between the calculated feature amount and the teacher data of the reference music piece. The likelihood is calculated and the music likelihood is output to the comparison unit 41. The music teacher data may be, for example, the music MFCC. Since the music likelihood calculating unit 27 calculates and outputs one music likelihood in a predetermined period, the music likelihood calculating unit 27 outputs the music likelihood at predetermined time intervals.

  The radio wave intensity acquisition unit 29 acquires the radio wave intensity output from the tuner unit 1. The radio wave intensity acquisition unit 29 outputs the acquired radio wave intensity to the comparison unit 41.

  The program information acquisition unit 31 acquires the broadcast station selected by the tuner unit 1 and the program information. For example, program information included in broadcast radio waves may be acquired from the tuner unit 1 or program information input by the user may be acquired. The program information includes channel information for identifying a broadcasting station and program information for identifying a program. The program information acquisition unit 31 outputs the acquired program information to the comparison unit 41.

  The music length acquisition unit 33 measures the time of the music content input from the music content extraction unit 21, and acquires the time from the start to the end of the music content as the music length. The music length acquisition unit 33 outputs the acquired music length to the comparison unit 41.

  The comparison unit 41 includes an evaluation index determination unit 43 that determines a plurality of evaluation indexes, a quality level calculation unit 45 that calculates a quality level, and a quality level comparison unit 47 that compares the quality levels.

  The evaluation index determination unit 43 includes a cut point evaluation index determination unit 51, a noise amount evaluation index determination unit 52, a noise ratio evaluation index determination unit 53, a music amount evaluation index determination unit 54, and a music ratio evaluation index determination unit 55. And a radio wave intensity evaluation index determination unit 56, a program evaluation index determination unit 57, and a time length evaluation index determination unit 58.

  The cut point evaluation index determination unit 51 receives the number of cut points from the cut point detection unit 23. The cut point evaluation index determination unit 51 determines a predetermined cut point evaluation index according to the number of input cut points, and outputs the determined cut point evaluation index to the quality level comparison unit 47. The cut point is a point in time when the amount of change in the sound power becomes a predetermined value or more. In the music content, the point in time when the amount of change in the audio power is equal to or greater than a predetermined value is a place where severe noise is locally included due to a surge or the like when a radio broadcast is received. For this reason, if there are many cut points, there is a high possibility that the music content extracted by the music content extraction unit 21 contains noise. Therefore, the cut point evaluation index becomes smaller as the number of cut points is larger.

  The noise amount evaluation index determination unit 52 receives a plurality of noise likelihoods from the noise likelihood calculation unit 25 at predetermined time intervals. The noise amount evaluation index determination unit 52 determines a maximum value of a plurality of noise likelihoods, determines a predetermined noise amount evaluation index corresponding to the maximum noise likelihood, and uses the determined noise amount evaluation index as a quality The data is output to the level comparison unit 47. The noise likelihood indicates the probability that the frequency of the music content extracted by the music content extraction unit 21 is similar to the frequency of the noise. Therefore, if the maximum noise likelihood is high, the noise likelihood is set to the noise frequency in the music content. The most similar part shows the degree of similarity to the noise frequency. For this reason, the noise amount evaluation index becomes smaller as the maximum noise likelihood becomes larger.

  The noise ratio evaluation index determination unit 53 receives a plurality of noise likelihoods from the noise likelihood calculation unit 25 at predetermined time intervals. The noise amount evaluation index determination unit 52 calculates an average value of a plurality of noise likelihoods, determines a predetermined noise ratio evaluation index corresponding to the average noise likelihood, and uses the determined noise ratio evaluation index as a quality The data is output to the level comparison unit 47. The noise likelihood indicates the probability that the frequency of the music content extracted by the music content extraction unit 21 is similar to the frequency of the noise. Therefore, if the average value of the noise likelihood is high, the noise likelihood is similar to the frequency of the noise in the entire music content. Indicates the degree to do. For this reason, the noise ratio evaluation index becomes smaller as the average value of noise likelihood becomes larger.

  The music amount evaluation index determination unit 54 receives a plurality of music likelihoods from the music likelihood calculation unit 27 at predetermined time intervals. The music amount evaluation index determination unit 54 determines the maximum value of a plurality of music likelihoods, determines a predetermined music amount evaluation index corresponding to the maximum music likelihood, and uses the determined music amount evaluation index as a quality. The data is output to the level comparison unit 47. The music likelihood indicates the likelihood that the frequency of the music content extracted by the music content extraction unit 21 is similar to the frequency of the music. Therefore, if the maximum value of the music likelihood is high, the portion most similar to the frequency of the music is displayed. Indicates the degree of similarity to the frequency of the music. For this reason, the music volume evaluation index becomes larger as the maximum value of the music likelihood is larger.

  The music ratio evaluation index determination unit 55 receives a plurality of music likelihoods from the music likelihood calculation unit 27 at predetermined time intervals. The music amount evaluation index determination unit 54 calculates an average value of a plurality of music likelihoods, determines a music ratio evaluation index determined in advance corresponding to the average music likelihood, and uses the determined music ratio evaluation index as a quality. The data is output to the level comparison unit 47. The music likelihood indicates the likelihood that the frequency of the music content extracted by the music content extraction unit 21 is similar to the frequency of the music, so that if the average value of the music likelihood is high, it is similar to the frequency of the music in the entire music content Indicates the degree to do. For this reason, a music ratio evaluation index becomes a larger value, so that the average value of music likelihood is large.

  The radio wave strength evaluation index determination unit 56 receives the telephone strength from the radio wave strength acquisition unit 29. The radio wave intensity while the broadcast radio wave including the music content is received by the tuner unit 1 is input. The radio field strength evaluation index determination unit 56 determines a predetermined radio field strength evaluation index with respect to the maximum value or average value of the input radio field strength, and outputs the determined radio field strength evaluation index to the quality level comparison unit 47. . If the radio field intensity is high, the probability that the music content includes noise is low. Therefore, the radio field intensity evaluation index increases as the maximum value or the average value of the radio field intensity increases.

  The program evaluation index determination unit 57 receives program information from the program information acquisition unit 31. The program evaluation index determination unit 57 determines a predetermined program evaluation index corresponding to the input program information, and outputs the determined program evaluation index to the quality level comparison unit 47. Features appear for each piece of music content depending on the broadcasting station or program. For example, music content included in FM radio broadcasting may include a song introduction by a disc jockey or the like at the beginning of the song, depending on the broadcast program. Depending on the user's preference, there is a possibility that a disc jockey likes or dislikes for each broadcast program (broadcast program). Therefore, the program evaluation index is an index for reflecting the user's intention, and is set for each program information by the user.

  The time length evaluation index determination unit 58 receives the music length from the music length acquisition unit 33, determines a predetermined time length evaluation index corresponding to the music length, and uses the determined time length evaluation index as a quality level calculation unit. Output to 45. Depending on the program program, the music content included in the FM radio broadcast may be broadcast only halfway through the music content. When the length of the original music content is known, the music content having a length close to it is preferable. The time length evaluation index becomes larger as it is closer to the original music length.

  The quality level calculation unit 45 includes a setting unit 61. The setting unit 61 sets a weighting coefficient for each of a plurality of evaluation indexes. If the user inputs the weighting coefficients for each of the plurality of evaluation indexes to the car audio device, the setting reception unit 35 receives the weighting coefficients for each of the plurality of evaluation indexes input to the operation unit 12. The setting receiving unit 35 outputs the weighting coefficient for each of the received plurality of evaluation indexes to the setting unit 61. The setting unit 61 stores the weighting coefficient input from the setting receiving unit 35 in the memory 8. Also, if the user inputs an operation for changing a plurality of weighting factors set in advance in the car audio device, the change is accepted, and the plurality of weighting factors set in advance and stored in the memory 8 are changed by the user. Rewrite with the weighted coefficient.

  The quality level calculation unit 45 includes a cut point evaluation index, a noise amount evaluation index, a noise ratio evaluation index, a music amount evaluation index, a music ratio evaluation index, a radio wave intensity evaluation index, a program evaluation index, and a program input from the evaluation index determination unit 43 The quality level is determined using the time length evaluation index and the weighting coefficient set by the setting unit 61, and the determined quality level and a plurality of evaluation indexes are output to the quality level comparison unit 47. Specifically, the quality level is calculated using the following equation (1).

Quality level Q = ΣWi × Ei (1)
(However, 1 ≦ i ≦ 8)
Here, E1 (Ei when i = 1) indicates a cut point evaluation index, and W1 (Wi when i = 1) indicates a weighting coefficient corresponding to the cut point evaluation index. E2 indicates a noise amount evaluation index, and W2 indicates a weighting coefficient corresponding to the noise amount evaluation index. E3 indicates a noise ratio evaluation index, and W3 indicates a weighting coefficient corresponding to the noise ratio evaluation index. E4 indicates a music amount evaluation index, and W4 indicates a weighting coefficient corresponding to the music amount evaluation index. E5 indicates a music ratio evaluation index, and W5 indicates a weighting coefficient corresponding to the music ratio evaluation index. E6 indicates a radio wave intensity evaluation index, and W6 indicates a weighting coefficient corresponding to the radio wave intensity evaluation index. E7 indicates a program evaluation index, and W7 indicates a weighting coefficient corresponding to the program evaluation index. E8 indicates a time length evaluation index, and W8 indicates a weighting coefficient corresponding to the time length evaluation index.

  The quality level comparison unit 47 compares the quality level input from the quality level calculation unit 45 with the quality level corresponding to the music content having the same content stored in the HDD 10. If the quality level input from the quality level calculation unit 45 is higher than the quality level corresponding to the music content of the same content stored in the HDD 10, the music content extracted by the music content extraction unit 21 is The corresponding cut point evaluation index, noise amount evaluation index, noise ratio evaluation index, music amount evaluation index, music ratio evaluation index, radio wave strength evaluation index, program evaluation index, and time length evaluation index are stored in the HDD 10. At this time, the music contents having the same contents stored in the HDD 10 to be compared and the corresponding evaluation index are deleted.

  In addition, when a user deletes the music content of the same content, you may make it provide a user with the value of each evaluation parameter | index as a judgment material.

  In addition, the quality level of each of the two music contents is set as a cut point evaluation index, a noise amount evaluation index, a noise ratio evaluation index, a music volume evaluation index, a music ratio evaluation index, a radio wave strength evaluation index, a program evaluation index, and a time length evaluation index. However, one of the eight evaluation indexes may be used as the quality level, or the quality level may be calculated from two or more counties selected from the eight evaluation indexes.

<Modification>
In the embodiment described above, the quality level is determined using a plurality of evaluation indexes, but the quality level can be determined using only the number of cut points. In this case, in FIG. 2, the noise likelihood calculation unit 25, the music likelihood calculation unit 27, the radio wave intensity acquisition unit 29, the program information acquisition unit 31, and the music length acquisition unit 33 are unnecessary. Furthermore, the quality level calculation unit 45 and the quality level comparison unit 47 are not necessary in the comparison unit 41, and the evaluation index determination unit 43 only has the cut point evaluation index determination unit 51. The comparison unit 41 determines the noise level of the music content extracted by the music content extraction unit 21 based on the cut point evaluation index determined by the cut point evaluation index determination unit 51. For example, it is determined whether or not to record using a threshold value.

  In the above-described embodiment, the car audio apparatus 1 has been described as an example of the music quality determination apparatus. However, the music quality determination method and the music quality determination method for executing the function illustrated in FIG. Needless to say, the present invention can be understood as a music quality determination program for causing a computer to execute.

  The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

<Appendix>
(1) The evaluation index determination means includes a cut point detection means for detecting, as a cut point, a time point when the audio signal level or the amount of change thereof is a predetermined value or more from the music content;
The music quality determination apparatus according to claim 1, further comprising: a cut point evaluation index determination unit that determines a cut point evaluation index based on the detected number of cut points.
(2) The evaluation index determination unit determines a radio field strength evaluation index based on a radio field intensity of the broadcast signal while the broadcast signal including the music content is received by the reception unit. The music quality determination apparatus according to claim 1, further comprising a determination unit.
(3) The evaluation index determination unit includes a program evaluation index determination unit that determines a program evaluation index based on a broadcast station name or a program name of the radio broadcast including the music content. The music quality determination apparatus according to any one of 1) to (2).
(4) The evaluation index determination unit includes a time length evaluation index determination unit that determines a time length evaluation index based on a time length of the music content. 3) The music quality determination apparatus according to any one of 3).
(5) receiving means for receiving broadcast radio waves;
Extracting means for extracting music content from the received broadcast radio wave;
A cut point detecting means for detecting, as a cut point, a time point when the audio signal level or the amount of change thereof is a predetermined value or more from the extracted music content;
A music quality judgment device comprising: judgment means for judging a noise level of the extracted music content based on the detected number of cut points.

When the audio signal level or the amount of change in the music content is equal to or higher than a predetermined value, there is a high possibility of noise, so that it is possible to provide a music quality determination device capable of determining the degree of noise.
(6) extracting music content from broadcast radio waves;
Storing the extracted music content;
Determining a plurality of evaluation indices for the extracted music content;
Calculating the quality level weighted to the plurality of evaluation indices;
When music content having the same title as the extracted music content is stored, the music quality determination includes the step of comparing the quality level of the extracted music content with the quality level of the stored music content Method.
(7) extracting music content from broadcast radio waves;
Detecting a point in time when the audio signal level or the amount of change thereof is a predetermined value or more from the extracted music content, as a cut point;
Determining a noise level of the extracted music content based on the detected number of cut points.

It is a figure which shows an example of the basic composition of the car audio apparatus in this Embodiment. It is a functional block diagram which shows an example of the outline | summary of the function which CPU has.

Explanation of symbols

  1 tuner unit, 2 A / D converter, 3 MP3 codec, 4 D / A converter, 5 speaker, 8 memory, 9 HDD interface unit, 11 external storage device, 12 operation unit, 21 music content extraction unit, 23 cut Point detection unit, 25 noise likelihood calculation unit, 27 music likelihood calculation unit, 29 radio wave intensity acquisition unit, 31 program information acquisition unit, 33 music length acquisition unit, 35 setting reception unit, 41 comparison unit, 43 evaluation index determination unit , 45 Quality level calculation section, 47 Quality level comparison section, 51 Cut point evaluation index determination section, 52 Noise amount evaluation index determination section, 53 Noise ratio evaluation index determination section, 54 Music volume evaluation index determination section, 55 Music ratio evaluation index A determination unit, 56 a radio wave strength evaluation index determination unit, 57 a program evaluation index determination unit, a 58 time length evaluation index determination unit, and a 61 setting unit.

Claims (6)

Receiving means for receiving broadcast radio waves;
Extracting means for extracting music content from the received broadcast radio wave;
Storage means for storing the extracted music content;
Comparing means for comparing the quality level of the extracted music content with the quality level of the stored music content when the same music content as the extracted music content is stored in the storage means ,
The comparison means includes an evaluation index determination means for determining a plurality of evaluation indices for each music content;
A music quality determination apparatus comprising: calculation means for calculating the quality level weighted to the plurality of evaluation indexes. The calculation means further includes setting means for setting a plurality of weighting coefficients corresponding to each of the plurality of evaluation indexes,
The music quality determination device according to claim 2, wherein when a change by the user is accepted, the plurality of weighting coefficients are set to values changed by the user. The evaluation index determining means is a noise likelihood calculating means for calculating a noise likelihood determined in relation to teacher data of noise determined in advance using a likelihood function for the music content;
The music quality determination apparatus according to claim 1, further comprising: a noise amount evaluation index determination unit that determines a noise amount evaluation index based on the calculated maximum value of the noise likelihood. The evaluation index determining means is a noise likelihood calculating means for calculating a noise likelihood determined in relation to teacher data of noise determined in advance using a likelihood function for the music content;
The music quality determination apparatus according to claim 1, further comprising: a noise ratio evaluation index determination unit that determines a noise ratio evaluation index based on the average value of the calculated noise likelihoods. The evaluation index determining means is a music likelihood calculating means for calculating a music likelihood determined by a relationship with teacher data of a music predetermined using a likelihood function for the music content;
The music quality determination apparatus according to claim 1, further comprising: a music volume evaluation index determination unit that determines a music volume evaluation index based on the calculated maximum value of the music likelihood. The evaluation index determining means is a music likelihood calculating means for calculating a music likelihood determined by a relationship with teacher data of a music predetermined using a likelihood function for the music content;
The music quality determination apparatus according to claim 1, further comprising: a music ratio evaluation index determining unit that determines a music ratio evaluation index based on the calculated average value of the music likelihoods.

Images