JP2005128355A - Karaoke singing ability rating system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To facilitate rating of Karaoke singing ability and to easily transfer Karaoke singing data. <P>SOLUTION: The Karaoke singing ability rating system 28 is equipped with a monaural signal conversion means 16 which converts the stereoscopic sound signals in which microphone input sound signals of a Karaoke singing person and the stereoscopic accompaniment signals of Karaoke playing are mixed to monaural sound signals, a spectrum analysis means 24 which performs the frequency spectrum analysis of the monaural sound signals, a dissonance detecting means 25 which detects the generation of the dissonance from the sound data subjected to the spectrum analysis, and a singing ability judging means 26 which judges the singing ability of the Karaoke singing person based on the generation of the dissonance. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  TECHNICAL FIELD The present invention relates to a karaoke singing ability scoring system, and in particular, karaoke singing ability that allows karaoke singing ability to be easily scored and karaoke singing recording data to be easily transferred for grading. It relates to the scoring system.

  Conventionally, a karaoke scoring device that is added to a karaoke device and scores the singing ability of a karaoke singer is well known, and the karaoke scoring device, for example, samples a voice signal of a singer input from a microphone. The pitches sung by the singer are extracted, the pitches are compared with the standard pitches corresponding to the main melody in the karaoke data, and a score corresponding to the comparison result is given for each appropriate scoring section. Upon completion, the scores are totaled to calculate a total score, and the total score is displayed on a scoreboard or display.

  Further, various karaoke scoring devices or karaoke scoring methods similar to the karaoke scoring device are known. For example, Patent Document 1 relates to the type of a reference audio signal such as an audio multiplex signal, a monaural signal, or a stereo signal. A pitch evaluation device and a karaoke scoring device capable of evaluating a voice signal of a singer are described, and the patent document 1 also describes a karaoke scoring device that allows a plurality of people to compare superiority of singing ability.

Patent Document 2 describes a singing scoring system that can obtain an appropriate scoring result in correspondence with singing by ad lib and singing by different versions.
JP 08-286684 A Japanese Patent Laid-Open No. 11-249664

  When finding talented karaoke singer nationwide and providing opportunities for professional debut, automatically compare the singer's singing voice with the standard melody as in the karaoke scoring system or singing scoring system above With the scoring system, you may overlook the broken corners. Therefore, it is desirable to record a singing voice along with a karaoke performance and have a professional teacher listen to it for review. However, only the primary screening should be processed automatically in order to narrow down a large number of applicants from all over the country. In that case, the screening center needs a device for separately processing the singing voice of the singer and the accompaniment music, and the system for extracting the reference melody for comparison with the singing voice of the singer from the accompaniment music is It is relatively complicated.

  In addition, when karaoke scoring / examination is performed at a place away from the karaoke device, there is a problem of data transfer. In that case, the transfer data can be reduced and the transfer can be easily performed. Is desired.

  Thus, there is a technical problem to be solved in order to make it easy to score karaoke singing power and to easily transfer karaoke singing data, and the present invention solves this problem. For the purpose.

  The present invention has been proposed in order to achieve the above-mentioned object, and the invention according to claim 1 includes a microphone input voice signal of a karaoke singer who performs karaoke singing in accordance with karaoke performance, and a stereo accompaniment signal of karaoke performance. A monaural signal converting means for converting a stereo audio signal mixed with a monophonic audio signal, a spectrum analyzing means for analyzing the frequency spectrum of the monaural audio signal, and a dissonance detecting means for detecting the occurrence of dissonance from the spectrum-analyzed sound data, The present invention provides a karaoke singing ability scoring system comprising singing ability judging means for judging the singing ability of a karaoke singer based on the occurrence of the dissonance.

  According to the first aspect of the present invention, since it is only necessary to analyze the frequency spectrum of the monaural audio signal, detect the occurrence of the dissonance, and determine the singing ability of the karaoke singer based on the occurrence of the dissonance, the singing ability is very easy. And the monaural audio signal can be easily transferred.

  Moreover, invention of Claim 2 is comprised so that the said singing power determination means may determine the singing power of a karaoke singer based on the said dissonance, the microphone input area of a karaoke singer, and the exemplary singing section of a karaoke song. The karaoke singing ability scoring system according to claim 1 is provided.

  According to the second aspect of the present invention, since it can be determined whether or not the singing section is to be sung based on the microphone input section of the karaoke singer and the exemplary singing section of the karaoke song, the determination of the singing ability is certain. become.

  Further, in the invention described in claim 3, the monaural signal converting means and the spectrum analyzing means are connected to each other through transmission data creating means and communication means, and the recording of the monaural audio signal created by the transmission data creating means is recorded. The karaoke singing ability scoring system according to claim 1 or 2, wherein the file is transferred to the spectrum analyzing means via the communication means.

  According to the third aspect of the present invention, the monaural signal generating means and the spectrum analyzing means can be separated, and the karaoke singing ability can be scored at a place away from the place where karaoke singing is performed.

  Furthermore, the invention according to claim 4 is configured such that the information of the microphone input section and the model song section is inserted into the header portion of the recording file and transferred. It provides a force scoring system.

  According to the fourth aspect of the present invention, information on the microphone input section and the model singing section can be easily transferred.

  The invention according to claim 5 is the karaoke singing ability according to claim 3, 4 or 5, wherein the recording file transferred through the communication means is a file in which a compressed audio signal obtained by compressing the monaural audio signal is recorded. A scoring system is provided.

  According to the fifth aspect of the present invention, the amount of transfer data can be reduced.

  The invention described in claim 6 provides the karaoke singing ability scoring system according to claim 6, wherein the compression is MP3 compression.

  According to the sixth aspect of the invention, the same effect as in the fifth aspect can be expected.

  Furthermore, the invention described in claim 7 is configured such that the personal ID (identification number) of the karaoke singer is inserted into the header portion of the recording file transferred through the communication means. , 4, 5 or 6 is provided.

  According to the seventh aspect of the invention, the karaoke singer can be easily identified and the data can be easily transferred.

  The invention according to claim 8 is configured such that the face photo data of the karaoke singer is inserted into the header portion of the recording file transferred through the communication means. , 6 or 7 karaoke singing ability scoring system is provided.

  According to the eighth aspect of the present invention, the face of the karaoke singer can be easily confirmed, and data can be transferred easily.

  The invention according to claim 1 of the present invention converts a stereo audio signal obtained by mixing a microphone input audio signal of a karaoke singer who performs karaoke singing in accordance with a karaoke performance and a stereo accompaniment signal of the karaoke performance into a monaural audio signal. Monaural signal conversion means, spectrum analysis means for frequency spectrum analysis of the monaural audio signal, dissonance detection means for detecting the occurrence of dissonance from the sound data subjected to spectrum analysis, and singing ability of a karaoke singer based on the occurrence of the dissonance The karaoke singing ability scoring system includes a singing ability judging means for judging, so that the monaural signal is subjected to frequency spectrum analysis based on the detected input time of the microphone, the occurrence of the dissonance is detected, and the karaoke is based on the occurrence of the dissonance. Since it is only necessary to determine the singing ability of the singer, it is extremely easy and quick. It is possible to determination of the proper singing force by. Also, by using a monaural signal, the amount of data transferred can be reduced and the karaoke singing ability scoring system can be simplified.

  Moreover, invention of Claim 2 is comprised so that it can determine whether the said singing power determination means was singing into the area which should be sung based on the microphone input area of a karaoke singer, and the model song area of karaoke music. Therefore, in addition to the effect of the invention of claim 1, the singing ability is reliably determined.

  Further, in the invention described in claim 3, the monaural signal converting means and the spectrum analyzing means are connected to each other through transmission data creating means and communication means, and the recording of the monaural audio signal created by the transmission data creating means is recorded. Since the file is configured to be transferred to the spectrum analysis means via the communication means, in addition to the effect of the invention according to claim 1 or 2, the monaural signal creation means and the analysis means can be separated from each other. The karaoke singing ability can be scored at a place away from the place where karaoke singing is performed.

  Furthermore, the invention described in claim 4 is configured such that the information on the microphone input section and the model singing section is inserted into the header portion of the recording file and transferred. In addition to the effects of the invention, information on the microphone input section and the model singing section can be easily transferred.

  In the invention according to claim 5, since the recording file transferred through the communication means is a file in which a compressed audio signal obtained by compressing the monaural audio signal is recorded, the effect of the invention according to claim 3 or 4 is achieved. In addition, the amount of transfer data can be reduced.

  Further, in the invention described in claim 6, since the compression is MP3 compression, the same effect as that of the invention described in claim 5 can be expected.

  Furthermore, the invention according to claim 7 is configured such that the personal ID (identification number) of the karaoke singer is inserted into the header portion of the recording file transferred through the communication means. In addition to the effect of the invention of item 3, 4, 5 or 6, the karaoke singer can be easily identified and the data can be transferred easily.

  The invention according to claim 8 is configured such that the face photo data of the karaoke singer is inserted into the header part of the recording file transferred through the communication means. , 5, 6 or 7, in addition to the effects of the invention of karaoke singer, it is easy to confirm the face of the karaoke singer, and the data can be easily transferred.

  For the purpose of making it easy to score karaoke singing power and to easily transfer karaoke singing data, a monaural mix signal that combines karaoke singers' singing signals and karaoke performance accompaniment signals is monaural. Realized by converting to a signal and compressing and transferring, analyzing the frequency spectrum of the monaural signal to detect the occurrence of dissonance, and determining the singing ability of a karaoke singer based on the occurrence of the dissonance .

  In the figure, reference numeral 1 denotes a stationary karaoke apparatus installed in a karaoke box or the like. The karaoke apparatus 1 signals a musical sound generation unit 2 that generates musical sounds as accompaniment to karaoke and a singing sound of a karaoke singer. The singing signal generating unit 3 for converting to singing, the sound outputting unit 4 for outputting sound, and the karaoke singing power scoring data creating unit 5 are configured.

  The musical tone generator 2 stores a music data memory 6 that stores a plurality of music data for playing karaoke music, a sequencer 7 that reads out the requested music data, and a karaoke of an orchestra or band based on the music data. A musical sound synthesizer 8 that synthesizes performance sounds, an effector 9 that applies effects such as hall reverberation to karaoke performance sounds, and a DAC (digital-analog converter) 10 that converts digital karaoke performance sounds into analog musical sounds Consists of Note that the tone synthesis unit 8, the effector 9, and the DAC 10 have a stereo configuration with two systems of a right channel and a left channel.

  The music data stored in the music data memory 6 is composed of a plurality of tracks and data for simultaneously generating musical sounds of a plurality of parts such as orchestras and bands. In addition to the performance track, the song data includes a guide melody track, a lyrics track, an effect control track, and the like, and song title information and song number information are written in the header.

  The singing signal generation unit 3 includes a microphone 11 for inputting a singing voice and an effector 12 for applying an effect such as an echo to the singing voice.

  Furthermore, the audio output unit 4 includes a mixer 13 for mixing the singing audio signal and the karaoke performance sound, an amplifier 14 for amplifying the sound, and a speaker 15. FIG. 4 shows details of the mixer 13 that mixes the karaoke performance sound output in stereo and the singing voice signal output in monaural and outputs in stereo. Therefore, the amplifier 14 and the speaker 15 are also configured in stereo.

  Further, the karaoke singing power scoring data creation unit 5 includes monaural signal conversion means 16 for converting the stereo output signal of the mixer 13 into a monaural audio signal, and details thereof are shown in FIGS. As shown in FIG. 5A, the audio signals of the left and right channels are mixed and converted to a monaural audio signal.

  Note that either one of the left and right channels may be selectively output as shown in FIG. In other words, a karaoke performance sound that has no room for dissonance is sufficient for reference, so only one channel can be substituted.

  Further, the karaoke singing power scoring data creation unit 5 detects the presence or absence of an input signal of the microphone 11 and detects a time section in which the microphone is input, and information on the guide melody track. The model singing section that can be detected from the above, that is, the model singing section detecting means 17a for detecting the time section that the singer should sing, the ID number input means 18 for inputting the ID number assigned to the singer, and the singing Singer's face photo creating means 19 for creating a person's face photo, and the transmission signal by combining output signals of the monaural signal converting means 16, microphone signal detecting means 17, ID number input means 18 and singer's face photo creating means 19 The transmission data creation means 20 created as a transmission / reception unit and a transmission / reception device which is a communication means for connecting to a public line, an Internet line, a dedicated line, or the like It consists of 21 Metropolitan. The transmission data creation means 20 includes data compression means 20a for compressing data into MP3 data. The transmission data may be transmitted in real time by a streaming method, or may be transmitted as a recording file recorded in a file format.

  Thus, when the user inputs the requested song to the sequencer 7, the sequencer 7 sequentially reads the requested song data from the song data memory 6 and inputs it to the musical tone synthesis unit 8. In parallel with this, the sequencer 7 sends the music data of the guide melody track to the model singing section detecting means 17a.

  The musical tone synthesis unit 8 synthesizes karaoke performance sounds of orchestras and bands in stereo format based on the input music data. The synthesized stereo karaoke performance sound is input to the effector 9, and the effector 9 gives an effect such as hall reverberation to the input karaoke performance sound. The above processes are all performed digitally, and the digital karaoke performance sound to which the effect is applied is converted into an analog musical sound by the DAC 10. The karaoke performance sound converted into an analog signal is output from the speaker 15 via the mixer 13 and the amplifier 14.

  The voice of the karaoke singer is input to the microphone 11, and an effect sound such as an echo is given by the effector 12, then input to the mixer 13, and mixed with the karaoke performance sound by the mixer 13. 14 from the speaker 15.

  Further, the karaoke singing power scoring data creation unit 5 converts the stereo audio output signal of the mixer 13 into a monaural audio signal by the monaural signal conversion means 16, and the microphone signal detection means 17 outputs the output signal of the microphone 11 To detect the section in which the singing sound is input to the microphone 11, output the microphone input section data represented by the input start time and the input end time, and the singer should sing by the exemplary singing section detection means 17 a Detecting the section, outputting the model singing section data representing the section by the start time and the end time, inputting the ID number of the karaoke singer by the ID number input means 18, and further, the singer face photo creation means 19 Then, the photograph of the face of the karaoke singer is created and these output signals are produced as transmission data by the transmission data creation means 20. To. That is, the microphone input section data and the model singing section data described above are recorded in the header portion of the recording file in which the monaural sound signal output from the monaural signal converting means 16 is sampled and AD-converted and recorded in the PCM format. Data and transmission data are created by inserting face photo data of the karaoke singer. Further, the monaural audio data in the transmission data is compressed into MP3 format audio data by the data compression means 20a. Of course, the compression method is not limited to MP3, and an appropriate audio compression method such as MP4 or twinVQ can be adopted.

  On the other hand, in the figure, 22 is a karaoke singing ability scoring unit, and the karaoke singing ability scoring unit 22 receives a recording file transmitted from the transmission / reception means 21 of the karaoke singing ability scoring data creation unit 5. A transmission / reception means 23, and a spectrum analysis means 24 for performing a frequency spectrum analysis on the PCM-format audio data obtained by decoding the compressed audio data of the recording file received by the transmission / reception means 23. The frequency spectrum analysis data includes a dissonance detecting means 25 for detecting a dissonance, and a singing ability determination means 26 for determining a singing ability based on the result of the detection of the dissonance. The dissonance detecting means 25 is connected to a consonance table 27 which stores a plurality of combinations of note names having a relation of consonances. The consonance table 27 stores all combinations of note names (consonance name groups) constituting the consonance. In order to determine the correspondence with the frequency obtained by the frequency analysis, a pitch name-frequency conversion table may be prepared separately, or the consonance table 27 itself may be a combination of frequencies constituting the consonance. Good. In any case, if all of the several frequencies detected simultaneously by the frequency analysis match one of the consonant name groups, it can be determined that no dissonance is included, and it matches any consonant name group. If not, there will be a mix of dissonances. Therefore, it can be determined that the spectrum of the frequency that does not match in the Kyodo group having the highest degree of matching is the one that the singer has pronounced incorrectly.

  The karaoke singing power scoring unit 22 and the karaoke singing power scoring data creation unit 5 constitute a karaoke singing power scoring system 28.

  Thus, when the transmission / reception means 23 receives the voice data transmitted from the transmission / reception means 21 of the karaoke singing power scoring data creation section 5, the karaoke singing power scoring section 22 receives the voice data by the spectrum analysis means 24. Frequency spectrum analysis is performed, the frequency spectrum analyzed by the dissonance detecting means 25 is collated with the consonance data of the consonance table 27 to detect dissonance, and the singing ability determination means 26 detects the dissonance. The singing ability is determined based on the result. The determination result may be sent from the transmission / reception means 23 to the transmission / reception means 21 and displayed on the display unit of the karaoke apparatus 1 or may be transmitted to the karaoke singer by another transmission means.

  FIG. 2 shows the waveform of the audio data received by the transmitting / receiving means 23 of the karaoke singing ability scoring unit 22, and the accompaniment sound waveform and the singing audio waveform are mixed in the waveform of the audio data.

  The result of frequency spectrum analysis of the sampled analysis portion shown in the figure by the spectrum analysis means 24 is the frequency spectrum distribution diagram shown in FIG.

  The analyzed frequency shown in the figure is collated with the consonance table 27 by the dissonance detecting means 25 to detect the dissonance shown in the figure. And the lower the frequency of occurrence of the detected dissonance, the higher the singing ability is scored. Further, the frequency spectrum analysis is repeated for a plurality of analysis parts to be sampled. When all of the voice data is analyzed and scored, the scores are totaled to calculate a total score.

  It is assumed that the karaoke performance sound does not include a dissonance, and if a dissonance is detected there, it means that the singing sound includes an incorrect pitch. This is the point to be deducted. However, if the singer does not sing where to sing, this should also be a big deduction. However, since a dissonance is not detected in that portion, it is overlooked.

  Therefore, by comparing the microphone input section data with the model singing section data, it is possible to determine whether or not the place to be sung is being sung. As shown in FIG. 6, the degree of coincidence between the section indicated by the model singing section data (the section to be sung) and the section indicated by the microphone input section data is also reflected in the singing ability scoring. This way, you can score reliably. Since the data indicates a time interval, the amount of data to be attached is small and the processing for comparing the intervals is simple.

  Thus, the karaoke singing ability scoring system can analyze the frequency spectrum of the monaural audio signal, detect the occurrence of the dissonance, and determine the singing ability of the karaoke singer based on the occurrence of the dissonance. In addition, an appropriate singing ability can be determined, and the monaural audio signal requires only half the amount of data compared to a stereo audio signal, and can be easily transferred.

  However, in this configuration, since it cannot be determined whether or not the place to be sung is being sung, the determination result lacks certainty. However, it is possible to pass through a rough screen without missing the masterpiece of a large number of audition applicants. It is effective as a primary examination to improve the efficiency of strict examinations conducted thereafter.

  Further, by adding a configuration for comparing the microphone input section and the model singing section, it is possible to determine whether or not a place to be sung is being sung.

  Furthermore, the monaural signal creation means and the spectrum analysis means can be separated so that the karaoke singing ability can be scored at a place away from the place where karaoke singing is performed.

  Furthermore, the microphone input section data and the model singing section data can be easily transferred.

  By using a monaural audio signal, the analysis means, detection means, and determination means can be simplified, and the amount of transfer data can be reduced.

  In addition, since the monaural audio signal is compressed into an MP3 compressed signal, the amount of transfer data can be further reduced.

  Further, since the personal ID of the karaoke singer is inserted into the monaural audio signal, the karaoke singer can be easily identified.

  Furthermore, since the face image data of the karaoke singer is inserted into the monaural audio signal, the karaoke singer's face can be easily confirmed.

  In the karaoke singing power scoring system 28, the karaoke singing power scoring data creation section 5 and the karaoke singing power scoring section 22 are provided in places separated via communication means. In this case, no communication means is required.

  In addition, the present invention can be variously modified without departing from the spirit of the present invention, and the present invention naturally extends to the modified one.

The explanation which shows one embodiment of the present invention and explains a karaoke singing ability scoring system Synthetic sound wave diagram Frequency spectrum distribution chart of synthesized sound Detailed conceptual diagram of mixer Conceptual diagram of monaural signal conversion Explanatory drawing about comparison between model singing section and microphone input section

Explanation of symbols

16 monaural signal converting means 17 microphone signal detecting means 24 spectrum analyzing means 25 dissonance detecting means 26 singing ability judging means 27 consonant table 28 karaoke singing ability scoring system

Claims (8)

  A monaural signal conversion means for converting a stereo audio signal obtained by mixing a microphone input audio signal of a karaoke singer who performs karaoke singing in accordance with a karaoke performance and a stereo accompaniment signal of the karaoke performance into a monaural audio signal, and the monaural audio signal Spectrum analysis means for frequency spectrum analysis, dissonance detection means for detecting the occurrence of dissonance from spectrum-analyzed sound data, and singing ability determination means for determining the singing ability of a karaoke singer based on the occurrence of the dissonance A karaoke singing ability scoring system.   2. The karaoke singing ability according to claim 1, wherein the singing ability determination means determines the singing ability of the karaoke singer based on the dissonance, the microphone input section of the karaoke singer, and the model singing section of the karaoke song. Scoring system.   The monaural signal converting means and the spectrum analyzing means are connected via transmission data generating means and communication means, and a recording file recording the monaural audio signal generated by the transmission data generating means is connected to the spectrum via the communication means. It is comprised so that it may be transmitted to an analysis means, The karaoke singing power scoring system of Claim 1 or 2 characterized by the above-mentioned.   4. The karaoke singing ability scoring system according to claim 3, wherein information on the microphone input section and the model singing section is inserted into a header portion of the recording file and transferred.   The karaoke singing ability scoring system according to claim 3 or 4, wherein the recording file transferred through the communication means is a file in which a compressed audio signal obtained by compressing the monaural audio signal is recorded.   6. The karaoke singing ability scoring system according to claim 5, wherein the compression is MP3 compression.   7. The personal ID (identification number) of the karaoke singer is inserted into the header portion of the recording file transferred through the communication means. The karaoke singing ability scoring system described.   8. The face image data of the karaoke singer is inserted into the header portion of the recording file transferred through the communication means. Karaoke singing ability scoring system.

Images