JP2010224013A - Karaoke machine stopping performance with voice uttered by singer to microphone - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To stop accompaniment music harmoniously and smoothly, midway through the music only with voice instructions to a microphone, while a singer who holds the microphone and is in the middle of the performance is not aware of such a troubles as voice feature parameter registration. <P>SOLUTION: The voice of the singer is analyzed, and feature parameters are extracted, based on singing voice signals obtained from the microphone 21 of the singer and character data in lyrics. Whether the singer who sang immediately prior utters words for performance stopping-instructions is analyzed, based on the feature parameter which a speech analysis means analyzed, until immediately prior and the words for performance stopping-instructions stored in a memory, during an interlude section of the Karaoke-accompanied music. The Karaoke-accompanied music performed is stopped, midway through the music when a voice recognition means analyzes that the words for performance stopping-instructions are uttered. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a karaoke apparatus in which a singer ceases performance by voice uttered by a microphone, and more particularly to a karaoke apparatus in which karaoke accompaniment music being played can be terminated halfway by a singer's voice instruction.

  The karaoke device generates and reproduces accompaniment music of a song that is freely selected by the user, generates time-series lyric character data synchronized with the progress of the accompaniment music, and overlays this on the background video to display (For example, Patent Documents 1 and 2).

  This karaoke device is often used in groups of a plurality of people. For example, one karaoke device installed in each room of a karaoke box is often shared by a plurality of customers in each room.

  For this reason, one karaoke apparatus is reserved for performance by a plurality of customers, and a queue for performance processing is made. Since it is possible to accept a performance reservation at any time, a queue is created even if there is only one customer. When there is no queue, the karaoke apparatus is in a standby state (idling state) until the next performance reservation is made, but this standby is often avoided because it wastes time and fees. Normally, it is always used with a queue.

  In addition, when a single karaoke device is used by a plurality of people, it is desirable that all of the plurality of users have an equal opportunity to sing while switching appropriately. It is necessary for each user to practice efficient use that makes money.

  In the karaoke device, the song of the song reserved by the user is sung along with the accompaniment music, but the singer does not always sing until the performance of the accompaniment music is finished, for example, only the first lyrics Singing the second and later lyrics may be desired to skip the singing and advance the next reserved song. Alternatively, there may be a case where only the favorite phrase portion of the lyrics is sung and the next reservation is to be played. Especially in young people's karaoke use scenes, they often request a song that they want to sing, and when the turn comes, sing to the point where they want to sing, stop playing immediately, and immediately start playing the next requested song. It can be seen.

  When the accompaniment music being played is terminated halfway, if the operation is performed in the interlude section of the accompaniment music without singing, it will not be accompanied by much discomfort, that is, without impairing the atmosphere of the place, as a natural flow The accompaniment music can be terminated halfway, and the next reserved song can be advanced.

  In this case, the singer is the best person who can complete the accompaniment music smoothly and smoothly. If you are a singer, you can freely determine the pros and cons of the end of the accompaniment and its end timing, but you may not be able to operate it at a suitable time with a third party other than the singer, Or it is because it is concerned that it will operate contrary to a singer's will.

  Even if the remote control can be done with the remote control, if there is no remote control at hand of the singer who wanted to cancel the performance, you can also ask the friend who has the remote control to stop the performance verbally is there. At this time, there is a situation in which information is mistakenly transmitted within the group, and the group stops performing even though it is still singing.

  However, it is quite cumbersome and difficult for the singer who is in a singing posture with the microphone in hand to perform the operation of terminating the accompaniment music at an appropriate timing.

  Therefore, a technique has been proposed in which a karaoke apparatus can be operated with a microphone (Patent Document 3). Patent Document 3 describes the technology of a wireless microphone with a voice recognition remote control function. However, if the microphone is operated with the remote control function of this microphone, a singer in a singing posture with the microphone in hand. Accompaniment music can be terminated halfway at any time without any difficulty.

JP 2004-317923 A JP2004-4896 JP2002-062894

  The wireless microphone with a voice recognition remote control function disclosed in Patent Document 3 performs a process for recognizing a voice acoustically input to the wireless microphone, and recognizes that a specific word is input, and associates it with the word. The stored code is modulated into an appropriate radio signal and transmitted, and all necessary components are built in the wireless microphone. In other words, it was completely independent from the karaoke device and completed only within the wireless microphone.

Here, there are two types of speech recognition: specific speaker speech recognition that recognizes only the speech of a specific speaker, and unspecified speaker speech recognition that recognizes the speech of an unspecified number of speakers. .
In the specific speaker voice recognition, input voice recognition processing is performed using the voice feature parameters of a specific speaker stored and registered in a memory in advance. This specific speaker voice recognition has the trouble of having to register the voice feature parameters of the user who is the recognition target in advance, but instead has the advantage of high recognition accuracy and few malfunctions even in environments with high background sounds. There is.
On the other hand, unspecified speaker voice recognition is not limited to a speaker, and any voice can be recognized, and there is no hassle of registering a user's voice feature parameters in advance. Therefore, it is suitable for performing speech recognition in one complete device system, such as a wireless microphone with a speech recognition remote control function in Patent Document 3, for example. Instead, the recognition accuracy is greatly inferior to that of specific speaker speech recognition, and there is a problem that malfunction is likely to occur in an environment with a large background sound.

The karaoke site where voices such as singing voices, talking voices and cheers constantly fly is inferior as an environment for voice recognition. It is ineligible and needs to be at least specific speaker speech recognition.
However, in a karaoke device used by an unspecified number of people, the specific speaker voice recognition that limits the speaker does not match the actual usage of the karaoke device. In order to perform this, it is necessary to force each user to fully understand the specific speaker voice recognition function, and to force the user to register the voice feature parameters in advance for each user.

  The present invention has been made in view of the technical background peculiar to karaoke as described above. The purpose of the present invention is that a singer who is in a singing posture with a microphone is aware of the trouble of registering voice feature parameters. Without being impaired, the accompaniment music can be smoothly and smoothly terminated by simply giving a voice instruction to the microphone. An object of the present invention is to provide a karaoke apparatus capable of moving a reservation-waiting song forward.

  The karaoke apparatus according to the present invention is specified by the following items (1) to (4).

(1) It is a karaoke device provided with voice analysis means, voice recognition means, and performance stop means. (2) The voice analysis means is a singing voice signal obtained from a microphone of a singer who sings along with the progress of karaoke accompaniment music. And analyzing the voice of the singer based on the time-series lyric character data synchronized with the progress of the accompaniment music, and extracting feature parameters. (3) The voice recognition means Based on the characteristic parameters analyzed by the voice analysis means until immediately before and the performance stop instruction word stored in the memory, it is analyzed whether or not the singer who was singing just before has uttered the performance stop instruction word (4) The performance stopping means, when analyzing that the voice recognition means has uttered the performance instruction word, ends the karaoke accompaniment music being played halfway.

  In a karaoke device, a singer who is in a singing posture with a microphone in hand quits accompaniment music smoothly and smoothly with just a voice instruction to the microphone without worrying about the trouble of registering voice feature parameters. be able to. As a result, for example, the next reservation-waiting music can be advanced and played as a natural flow without impairing the lively atmosphere of the karaoke site.

It is a functional block diagram of the karaoke apparatus which comprises one Example of this invention. It is a block diagram which shows the outline of the function added to the central processing unit, and its operation | movement.

=== Basic configuration and operation of karaoke apparatus ===
A schematic configuration of a karaoke apparatus according to an embodiment of the present invention is illustrated in FIG.
This karaoke apparatus is a computer application device equivalent to a well-known personal computer, and the central processing unit 11 that forms the core of the karaoke apparatus forms a computer main body including a CPU, a RAM, and a ROM.

  Under the control and management of the central processing unit 11, a hard disk device 12 as a large-capacity external storage, an optical disk playback device 13 such as a CD-ROM or DVD-ROM, and a karaoke host device via a public communication line such as an optical communication line A communication control device 14 for communication, a user interface device 15 for exchanging input from a user and a response to the user, and a music generation device 16 for generating an acoustic signal of accompaniment music based on MIDI music performance data , An acoustic device 17 that amplifies an accompaniment music and an acoustic signal from the microphone 21 and generates sound from the speaker 22, a display 18 using an LCD, a PDP, etc., a video processing device 19 that processes video data to be displayed on the display 18, etc. Is installed.

  The hard disk device 12 stores, for many karaoke songs, karaoke data including accompaniment music data mainly composed of MIDI data and lyric character data that is a generation origin of the lyric image. In addition, for the long-format video data in a predetermined format, script data that defines the video data processing sequence (storage location and processing order of video data to be processed, etc.), and karaoke songs that can be played, song names and artist names Table of contents information such as the release year and the singing part of the lyrics is also stored.

  The central processing unit 11 identifies karaoke data, script data, and table of contents information of each music piece by music number, and manages this as a karaoke database.

  When receiving the performance reservation command from the user interface device 15, the central processing unit 11 registers the music ID (music identification code) included in the received command in the performance reservation queue in the order received. Then, the music IDs are extracted from the queue in the order of registration, and the karaoke data for the corresponding music is extracted from the karaoke database and used for performance processing.

  The music generation device 16 generates accompaniment music from the accompaniment music data in the karaoke data. As for the lyric character data, the lyric image in which the portion to be sung is changed in color in synchronization with the accompaniment music is successively developed in the video RAM as a bitmap. Also, predetermined moving image data is sequentially transferred to the video processing device 19 in a predetermined order based on the script data, and the background moving image of the lyrics image is decoded.

  The acoustic device 17 includes a mixing amplifier, and mixes and amplifies the accompaniment music generated by the music generation device 16 and the singing voice input to the microphone 22 and outputs the sound from the speaker 22. The acoustic device 17 includes an AD converter 171 for digitally processing the audio signal input from the microphone 21.

  The video processing device 19 superimposes the lyrics image on the decoded moving image and displays it on the display 18. Lyric images are created sequentially from time-series lyric character data synchronized with the progress of accompaniment music.

  The user interface device 15 includes an operation panel of the karaoke device main body and a karaoke remote control device, and includes short-range wireless communication means (IrDA transceiver, infrared LED, infrared light receiving element) capable of bidirectional communication.

=== Functions Added to the Central Processing Unit 11 and their Operations ===
In the karaoke apparatus according to the present invention, the operation of the entire karaoke apparatus is performed under the control and management of the central processing unit 11, and the central processing unit 11 includes voice analysis, voice recognition, and interlude as shown in FIG. Functional parts such as section discrimination and performance stop are added in software.

  The voice analysis function unit sings along with the progress of the karaoke accompaniment music when the music generation device 16 generates the accompaniment music and the lyrics character data is sequentially output in synchronization with the accompaniment music. On the basis of the singing voice signal obtained from the person's microphone 21 and the time-series lyric character data synchronized with the progress of the accompaniment music, the singer's voice is analyzed to extract feature parameters.

  In this singing section, by analyzing the voice of the singer in correspondence with the lyric characters to be sung, characteristic parameters necessary for the voice recognition of the singer are extracted. The extracted feature parameters are accumulated while being sequentially updated in the memory. Only the latest feature parameters are always stored and stored in the memory. Further, when there are a plurality of common lyric characters, averaged characteristic parameters may be stored and stored.

  The singing section and the interlude section are discriminated based on the time data included in the karaoke data about the music being played, the time-series output pattern of the lyric character data, or the like.

  The voice recognition function unit analyzes the utterance of the singer obtained from the microphone 21 by the voice analysis function unit immediately before the interlude section when the accompaniment music continues but the lyric character data is not output. Speech recognition is performed based on the feature parameters stored in the memory. Based on the result of the voice recognition and the performance stop instruction word stored in advance in the memory, the singer who has sung until just before the interlude section performs the performance stop instruction words (for example, “Owari” “Chuushi” “ Tsugi, “stop”, “neck strike”, etc.).

  Here, when it is analyzed that the performance stop instruction word is uttered, the analysis result is transmitted from the voice recognition function unit as a performance stop command. The central processing unit 11 executes a procedure for ending the accompaniment music in response to the performance stop command, and if there is a next reserved waiting piece, the reserved waiting piece is advanced.

  If it is analyzed that the performance stop instruction word has not been spoken, the performance is continued.

  As described above, in the karaoke apparatus according to the present invention, the singer himself who is in a singing posture with the microphone in his / her hand can end the accompaniment music in a complete and smooth way with only a voice instruction to the microphone. However, at this time, the voice instruction of the singer himself / herself is recognized based on the characteristic parameter analyzed and extracted from the singing pronunciation during the singing by the singer.

  Thereby, the specific speaker voice recognition is performed with the singer limited to the speaker, and it is analyzed whether or not the singer has pronounced the performance stop instruction word based on the recognition result. This specific speaker voice recognition has the advantage of higher recognition accuracy and fewer malfunctions even in a karaoke environment where the background sound is large compared to unspecified speaker voice recognition, but by limiting the voice recognition to the interlude section. The performance stop instruction by the voice of the singer is more reliably performed.

  In order to recognize a specific speaker, a troublesome preparatory operation for registering a voice characteristic parameter of the specific speaker is necessary. However, in the karaoke apparatus according to the present invention, the singer does not need to be aware of such trouble at all, and singing Feature parameters necessary for voice recognition of a specific speaker are extracted just by singing normally in a section, and stored and registered in a memory.

  Then, the accompaniment music can be completed smoothly and smoothly in the middle just by singing a voice instruction to the microphone during the interlude section. Thereby, for example, the next reservation waiting song can be advanced and played as a natural flow without impairing the lively atmosphere of the karaoke site.

DESCRIPTION OF SYMBOLS 11 Central processing unit 12 Hard disk device 13 Optical disk reproducing device 14 Communication control device 15 User interface device 16 Music generating device 17 Audio device 171 AD converter 18 Display 19 Video processing device 21 Microphone 22 Speaker

Claims (1)

A karaoke apparatus comprising speech analysis means, speech recognition means, and performance stop means,
The voice analysis means analyzes the voice of the singer based on the singing voice signal obtained from the microphone of the singer singing along with the progress of the karaoke accompaniment music and the time-series lyric character data synchronized with the progress of the accompaniment music. Extract feature parameters,
The voice recognizing means is configured such that in the interlude section of karaoke accompaniment music, the singer who has been singing just before is based on the characteristic parameters analyzed by the voice analyzing means until immediately before and the performance stop instruction word stored in the memory. Analyzing whether or not the word for instructing to stop playing is spoken,
The performance canceling means is a karaoke device that terminates the karaoke accompaniment music being played when the voice recognition means analyzes that the performance instruction word has been uttered.

Images