JP2009092977A - In-vehicle device and music piece retrieval system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a music piece retrieval system 10 capable of performing retrieval and playback of a music piece including lyrics which is hummed by a user. <P>SOLUTION: The music piece retrieval system 10 stores a lyrics data for indicating the lyrics of the music piece for each music piece beforehand, recognizes a word string from voice of the music piece which is hummed by the user, and extracts and play back the music piece data which is related to the lyrics data including the word string. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an in-vehicle device that is mounted on a moving body such as a vehicle and has a function of reproducing music from music data.

  Patent Document 1 discloses a playback device that searches and plays back music data stored in a hard disk drive or the like based on an artist name, an album name, and a song name input by voice.

JP 2005-78705 A

  By the way, when the number of songs stored in a hard disk drive or the like increases, it may be difficult to remember the artist name, album name, and song name for each song. For this reason, in the technique disclosed in Patent Document 1, the user may not be able to reproduce a song for which the artist name, album name, and song name are forgotten, although the lyrics are remembered.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to search for and reproduce music containing lyrics that the user uttered.

  In order to solve the above-described problem, the in-vehicle device of the present invention stores, for each piece of music data, lyric data indicating the lyrics of the music in advance, recognizes a word string from the voice of the music spoken by the user, and the word string The music data corresponding to the lyrics data including is extracted and reproduced.

  For example, the first aspect of the present invention is an in-vehicle device mounted on a vehicle, and for each piece of music data, music information storage means for storing lyrics data indicating the lyrics of the music, and a record for storing music data Music data acquisition means for acquiring music data from the medium and storing it in the music information storage means, and lyrics data corresponding to the music data acquired by the music data acquisition means are acquired from outside and stored in the music information storage means The lyric data acquisition means, the speech recognition means for specifying the word string indicated by the voice from the voice uttered by the user, and the music information storage means, the same word string as the word string specified by the voice recognition means Music data extraction means for identifying lyrics data to be included and extracting music data associated with the identified lyrics data, and extracted by the music data extraction means Providing a vehicle device, characterized in that it comprises a reproducing device for reproducing the music data.

  The second aspect of the present invention is a music search system including an in-vehicle device mounted on a vehicle, and a lyrics data storage server that is provided outside the in-vehicle device and stores lyrics data for each song, The in-vehicle device acquires, for each piece of music data, music information storage means for storing lyrics data indicating the lyrics of the music, and music data acquisition for acquiring music data from a recording medium for storing the music data and storing it in the music information storage means Means, lyric data acquisition means for acquiring lyric data corresponding to the music data acquired by the music data acquisition means from the lyrics data storage server and storing it in the music information storage means, and the voice from the voice spoken by the user A speech recognition unit that identifies a word string to be shown, and a song data storage unit, and a lyric data that includes the same word string as the word string identified by the speech recognition unit A music data extraction means for extracting music data associated with the specified lyrics data, and a playback means for playing back the music data extracted by the music data extraction means Search system

  According to the in-vehicle device of the present invention, it is possible to search for and reproduce music containing lyrics sung by the user.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a system configuration diagram showing the configuration of a music search system 10 according to an embodiment of the present invention. The music search system 10 includes a lyrics data storage server 20, an audio data storage server 30, and an in-vehicle device 40. The lyrics data storage server 20 and the voice data storage server 30 are connected to a communication line 11 such as the Internet. The in-vehicle device 40 is mounted on the vehicle 13 and communicates with the lyrics data storage server 20 and the voice data storage server 30 via the communication line 11 by wirelessly communicating with the base station 12 connected to the communication line 11. .

  As shown in FIG. 2, the lyrics data storage server 20 includes a lyrics data storage unit 21 and a lyrics data transmission unit 22. For example, as shown in FIG. 3, the lyrics data storage unit 21 includes a song title 210, a singer name 211 singing a song corresponding to the song title 210, and an album in which the song corresponding to the song title 210 is recorded. The lyrics data 213 indicating the lyrics of the music corresponding to the song name 210 is stored in association with the album name 212.

  When the lyric data transmission unit 22 receives a lyric data acquisition request including the song name, singer name, and album name from the in-vehicle device 40 via the communication line 11, the lyric data transmission unit 22 Lyric data corresponding to the singer name and album name is extracted from the lyrics data storage unit 21. Then, the lyrics data transmitting unit 22 returns a lyrics data acquisition response including the extracted lyrics data to the transmission source of the lyrics data acquisition request.

  As shown in FIG. 4, the audio data storage server 30 includes a feature data storage unit 31 and a feature data transmission unit 32. For example, as shown in FIG. 5, the feature data storage unit 31 stores information indicating the feature amount 311 of the speech when the word 310 is uttered. When the feature data transmission unit 32 receives a feature value acquisition request including information indicating a word from the in-vehicle device 40 via the communication line 11, the feature data transmission unit 32 corresponds to the word included in the feature value acquisition request. Is extracted from the feature data storage unit 31, and a feature amount acquisition response including information indicating the extracted feature amount is returned to the transmission source of the feature amount acquisition request.

  For example, as shown in FIG. 6, the in-vehicle device 40 includes a likelihood calculating unit 41, a word string specifying unit 42, a music data extracting unit 43, a reproducing unit 44, a feature amount acquiring unit 45, a dictionary data storage unit 46, and lyrics data acquisition. Unit 47, music information storage unit 48, and music data acquisition unit 49.

  For example, as illustrated in FIG. 7, the dictionary data storage unit 46 stores information indicating a voice feature amount 461 when the word 460 is uttered for each word 460. For example, as shown in FIG. 8, the song information storage unit 48 includes a song name 480, a singer name 481 singing a song corresponding to the song name 480, and an album containing songs corresponding to the song name 480. In association with the album name 482, song data 483 of the song corresponding to the song name 480 and lyrics data 484 indicating the lyrics of the song corresponding to the song name 480 are stored.

  The music data acquisition unit 49 acquires music data from a recording medium 19 such as a CD (Compact Disc) when the user gives an instruction to acquire music via the input device 15 such as a touch panel. And the music data acquisition part 49 acquires the music name, singer name, and album name of the music corresponding to the said music data from the information input from the user or the information previously acquired from CDDB (Compact Disc DataBase) etc., The music data acquired from the recording medium 19 is stored in the music information storage unit 48 in association with the acquired music name, singer name, and album name. Then, the music data acquisition unit 49 notifies the lyrics data acquisition unit 47 that a new music has been registered.

  The lyric data acquisition unit 47 displays a screen for inquiring the user as to whether or not the lyric data can be acquired, such as an LCD (Liquid Crystal Display), when the music data acquisition unit 49 is notified that a new music has been registered. Is displayed on the display device 18.

  When the input indicating that the acquisition of the lyrics data is permitted is received from the user via the input device 15, the lyrics data acquisition unit 47 refers to the music information storage unit 48 and acquires the lyrics data. Music data that has not been specified is specified, a lyric data acquisition request including a tune name, a singer name, and an album name corresponding to the tune data is generated, and the generated lyric data acquisition request is transmitted via the antenna 17 to the lyric data storage server 20. Send to.

  When the lyrics data acquisition response including the lyrics data is received via the antenna 17, the lyrics data acquisition unit 47 associates the lyrics data included in the lyrics data acquisition response with the corresponding music data. It is stored in the music information storage unit 48.

  Then, the lyrics data acquisition unit 47 refers to the dictionary data storage unit 46 to determine whether or not all the words in the lyrics data included in the lyrics data acquisition response are stored in the dictionary data storage unit 46. judge. If there is a word that is not stored in the dictionary data storage unit 46 in the lyrics data included in the lyrics data acquisition response, the lyrics data acquisition unit 47 sends information indicating the word to the feature amount acquisition unit 45. .

  When the feature amount acquisition unit 45 receives information indicating a word from the lyrics data acquisition unit 47, the feature amount acquisition unit 45 generates a feature amount acquisition request including the word, and stores the generated feature amount acquisition request via the antenna 17 as voice data. Send to server 30. When the feature amount acquisition response including information indicating the feature amount is received via the antenna 17, the feature amount acquisition unit 45 corresponds to the information indicating the feature amount included in the feature amount acquisition response. It is stored in the dictionary data storage unit 46 together with the word.

  When the user is instructed to start a music search by voice via the input device 15, the likelihood calculation unit 41 takes in a voice signal through the microphone 14 and extracts a feature value for each syllable from the fetched voice signal. To do. Then, the likelihood calculating unit 41 refers to the dictionary data storage unit 46, compares the extracted feature amount with the feature amount of each word in the dictionary data storage unit 46, and the feature amount is similar. The likelihood that shows a higher value as the percentage of the number of words that are present is calculated for each word.

  Then, the likelihood calculating unit 41 outputs the calculated likelihood to the word string specifying unit 42 together with information indicating the corresponding word. Note that the likelihood calculating unit 41 outputs a predetermined number (for example, within 10) of words to the word string specifying unit 42 in the descending order of likelihood for each syllable together with the corresponding likelihood. Good.

  Based on the likelihood output from the likelihood calculating unit 41, the word string specifying unit 42 specifies, for example, the word having the highest likelihood as the word corresponding to the speech. Then, the word string specifying unit 42 outputs the plurality of specified word strings to the music data extracting unit 43. Note that the word string specifying unit 42 also considers the context of the high likelihood word output from the likelihood calculation unit 41 immediately before and the high likelihood word output from the likelihood calculation unit 41 thereafter. The word corresponding to the voice may be specified.

  When the music data extraction unit 43 receives a word string from the word string identification unit 42, the music data extraction unit 43 refers to the music information storage unit 48 and extracts music data associated with the lyric data including the word string. The extracted music data is sent to the playback unit 44. The playback unit 44 plays back the music data received from the music data extraction unit 43 via the speaker 16.

  When a plurality of lyrics data including the word string output from the word string specifying unit 42 exists in the music information storage unit 48, the music data extracting unit 43 selects the song name, singer name, and album name of the plurality of songs. Is displayed on the display device 18, and the playback unit 44 plays back the music specified by the user via the input device 15.

  When the lyrics data including the word string output from the word string specifying unit 42 does not exist in the music information storage unit 48, the music data extraction unit 43 adds the lyrics data including a part of the word string to the lyrics data. The song name, singer name, and album name of the associated song data are displayed on the display device 18, and the playback unit 44 plays back the song data when playback is designated by the user via the input device 15. You may make it do.

  At this time, the music data extraction unit 43 uses the song name, singer name, and album name of the music data associated with the lyrics data including the word included in the word string output from the word string specifying unit 42 as the word. You may make it display on the display apparatus 18 for several music (for example, 5 music) in order with many words contained in the row | line | column.

  FIG. 9 is a flowchart showing an example of the operation of the music search system 10 at the time of music data acquisition. When an instruction to acquire music data is given from the user via the input device 15, the music search system 10 starts the operation shown in this flowchart.

  First, the music data acquisition unit 49 acquires music data from the recording medium 19 in which music data is stored, and stores the acquired music data together with the music name, singer name, and album name corresponding to the music data. The data is stored in the unit 48 (S100). Then, the music data acquisition unit 49 notifies the lyrics data acquisition unit 47 that a new music has been registered.

  Next, the lyric data acquisition unit 47 displays a screen for inquiring the user whether or not the lyric data can be acquired on the display device 18, and receives an input indicating that the acquisition of the lyric data is permitted via the input device 15. It is determined whether it has been received from the user (S101). When the input indicating that the acquisition of the lyrics data is permitted is not received from the user (S101: No), the music search system 10 ends the operation shown in the flowchart.

  When the input indicating that the acquisition of the lyrics data is permitted is received from the user (S101: Yes), the lyrics data acquisition unit 47 refers to the music information storage unit 48 and does not execute the acquisition of the lyrics data. The data is specified, a lyrics data acquisition request including the song name, singer name, and album name corresponding to the music data is generated, and the generated lyrics data acquisition request is transmitted to the lyrics data storage server 20 via the antenna 17.

  When the lyrics data acquisition response including the lyrics data is received via the antenna 17, the lyrics data acquisition unit 47 associates the lyrics data included in the lyrics data acquisition response with the corresponding music data. It is stored in the music information storage unit 48 (S102).

  Next, the lyric data acquisition unit 47 refers to the dictionary data storage unit 46 and determines whether or not all words in the lyric data included in the lyric data acquisition response are stored in the dictionary data storage unit 46. Is determined (S103). When all the words in the lyrics data included in the lyrics data acquisition response are stored in the dictionary data storage unit 46 (S103: Yes), the music search system 10 ends the operation shown in this flowchart.

  When there is a word that is not stored in the dictionary data storage unit 46 in the lyrics data included in the lyrics data acquisition response (S103: No), the lyrics data acquisition unit 47 displays information indicating the word as a feature amount. The data is sent to the acquisition unit 45. Then, the feature amount acquisition unit 45 generates a feature amount acquisition request including the word received from the lyrics data acquisition unit 47 and transmits the generated feature amount acquisition request to the audio data storage server 30 via the antenna 17.

  And the feature-value acquisition part 45 acquires the feature-value of a word by receiving the feature-value acquisition response containing the information which shows a feature-value via the antenna 17 (S104). Then, the feature amount acquisition unit 45 stores the information indicating the feature amount included in the feature amount acquisition response in the dictionary data storage unit 46 together with the corresponding word (S105), and the music search system 10 performs this flowchart. The operation shown in FIG.

  FIG. 10 is a flowchart showing an example of the operation of the music search system 10 when reproducing music data. When the start of music search by voice is instructed by the user via the input device 15, the music search system 10 starts the operation shown in this flowchart.

  First, the likelihood calculating unit 41 captures an audio signal via the microphone 14 and extracts a feature amount for each syllable from the acquired audio signal. Then, the likelihood calculating unit 41 refers to the dictionary data storage unit 46 and compares the extracted feature amount with the feature amount of each word in the dictionary data storage unit 46 to thereby determine the likelihood for each word. (S200), and the calculated likelihood is output to the word string specifying unit 42 together with information indicating the corresponding word.

  Next, the word string specifying unit 42 specifies a word for each syllable based on the likelihood output from the likelihood calculating unit 41 (S201). Then, the word string specifying unit 42 outputs the specified plurality of words to the music data extracting unit 43 as word strings.

  Next, the music data extraction unit 43 extracts music data associated with the lyrics data including the word string received from the word string specifying unit 42 from the music information storage unit 48 (S202), and the extracted music data Is sent to the playback unit 44. Then, the playback unit 44 plays back the song data received from the song data extraction unit 43 via the speaker 16 (S203), and the song search system 10 ends the operation shown in this flowchart.

  FIG. 11 is a hardware configuration diagram illustrating an example of a hardware configuration of a computer 60 that realizes the functions of the lyrics data storage server 20, the voice data storage server 30, or the in-vehicle device 40. The computer 60 includes a CPU (Central Processing Unit) 61, a RAM (Random Access Memory) 62, a ROM (Read Only Memory) 63, an HDD (Hard Disk Drive) 64, a communication device 65, an input / output interface (I / F) 66, And a media interface (I / F) 67.

  The CPU 61 operates based on a program stored in the ROM 63 or the HDD 64 and controls each unit. The ROM 63 stores a boot program executed by the CPU 61 when the computer 60 is started up, a program depending on the hardware of the computer 60, and the like. The HDD 64 stores a program executed by the CPU 61.

  The communication device 65 receives data from other devices via the communication line and sends the data to the CPU 61, and transmits the data generated by the CPU 61 to other devices via the communication line. The input / output interface 66 receives a signal from the input / output device and sends it to the CPU 61, and outputs data acquired from the CPU 61 to the input / output device. The CPU 61 controls the input / output device via the input / output interface 66, acquires a signal from the input / output device via the input / output interface 66, and sends the generated data to the input / output device via the input / output interface 66. Output.

  The media interface 67 reads a program or data stored in the recording medium 68 and provides it to the RAM 62. A program provided to the CPU 61 via the RAM 62 is stored in the recording medium 68. The program is read from the recording medium 68, installed in the computer 60 via the RAM 62, and executed by the CPU 61. The recording medium 68 is, for example, an optical recording medium such as a DVD (Digital Versatile Disk) or PD (Phase change rewritable disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory. Etc.

  When the computer 60 functions as the lyrics data storage server 20, a program installed and executed on the computer 60 causes the computer 60 to function as the lyrics data storage unit 21 and the lyrics data transmission unit 22.

  When the computer 60 functions as the audio data storage server 30, a program installed and executed on the computer 60 causes the computer 60 to function as the feature data storage unit 31 and the feature data transmission unit 32.

  Further, when the computer 60 functions as the in-vehicle device 40, the program installed in the computer 60 and executed is the likelihood calculating unit 41, the word string specifying unit 42, the music data extracting unit 43, and the reproducing unit 44. , Function amount acquisition unit 45, dictionary data storage unit 46, lyrics data acquisition unit 47, music information storage unit 48, and music data acquisition unit 49.

  The computer 60 reads these programs from the recording medium 68 and executes them. As another example, the computer 60 may acquire these programs by the communication device 65 via a communication line.

  The embodiment of the present invention has been described above.

  As is clear from the above description, according to the music search system 10 of the present embodiment, it is possible to search for and play music containing lyrics that the user uttered. In addition, since the music search system 10 searches for music based on the lyrics that the user does not squeeze instead of the melody that the user screams, the lyric of the desired music can be obtained even if the user is a timid user or a user who does not have a sense of rhythm. Can be searched and played back without specifying a song name or the like.

  In addition, this invention is not limited to above-described embodiment, Many deformation | transformation are possible within the range of the summary.

  For example, the in-vehicle device 40 may be configured as shown in FIG. In the example illustrated in FIG. 12, the in-vehicle device 40 includes a likelihood calculating unit 41, a word string specifying unit 42, a music data extracting unit 43, a reproducing unit 44, a feature amount acquiring unit 45, a lyrics data acquiring unit 47, and a music information storage unit. 48, a music data acquisition unit 49, a second dictionary data 50, a first dictionary data 51, and a command execution unit 52. Except for the points described below, in FIG. 12, the components denoted by the same reference numerals as those in FIG. 6 have the same or similar functions as those in FIG.

  The first dictionary data 51 stores the feature amount of each word used only when searching for music by voice. The command execution unit 52 stores the feature amount of each word used for speech recognition other than music search.

  The lyrics data acquisition unit 47 stores the lyrics data included in the lyrics data acquisition response in the music information storage unit 48 in association with the corresponding music data, and then refers to the first dictionary data 51 to It is determined whether or not all words in the lyrics data included in the lyrics data acquisition response are stored in the first dictionary data 51. When there is a word that is not stored in the first dictionary data 51 in the lyrics data included in the lyrics data acquisition response, the lyrics data acquisition unit 47 sends information indicating the word to the feature amount acquisition unit 45. send.

  When the feature amount acquisition unit 45 receives a feature amount acquisition response via the antenna 17, the feature amount acquisition unit 45 displays the information indicating the feature amount included in the feature amount acquisition response together with the corresponding word in the first dictionary data 51. To store.

  The likelihood calculating unit 41 receives a voice signal via the microphone 14 when normal voice recognition is instructed by the user via the input device 15, and extracts a feature amount for each syllable from the acquired voice signal, With reference to the second dictionary data 50, the extracted feature value is compared with the feature value of each word in the second dictionary data 50, and the likelihood of each word is calculated. Then, the likelihood calculating unit 41 outputs the calculated likelihood to the word string specifying unit 42 together with information indicating the corresponding word and information indicating normal speech recognition.

  On the other hand, when the user instructs the start of music search by voice via the input device 15, the likelihood calculating unit 41 extracts a feature quantity for each syllable from the voice signal captured via the microphone 14, and the first The extracted feature value is compared with the feature value of each word in the first dictionary data 51, and the likelihood of each word is calculated. Then, the likelihood calculating unit 41 outputs the calculated likelihood to the word string specifying unit 42 together with information indicating the corresponding word and information indicating that it is a music search.

  When the word string specifying unit 42 receives the word and likelihood from the likelihood calculating unit 41 together with information indicating that it is normal speech recognition, for example, the word string specifying unit 42 specifies the word with the highest likelihood as the word corresponding to the speech. The specified word is output to the command execution unit 52. The command execution unit 52 executes a command corresponding to the word output from the word string specifying unit 42.

  On the other hand, when a word and likelihood are received from the likelihood calculating unit 41 together with information indicating that the music search is performed, the word string specifying unit 42 specifies, for example, the word having the highest likelihood as the word corresponding to the voice. The plurality of identified words are output to the music data extraction unit 43 as a word string.

  In the above-described embodiment, the computer 60 acquires the lyrics data from the lyrics data storage server 20 via the communication line 11. However, as another form, the computer 60 inputs from the user via the input device 15. The text data or text data input via a recording medium such as a memory card may be stored in the music information storage unit 48 in association with music data designated by the user as lyrics data.

  In the embodiment described above, when the music data is acquired, the computer 60 acquires the corresponding lyrics data and the feature amount of the word from an external server. You may make it acquire the feature-value of a word in the case of transmission / reception of other data, such as at the time of update of map data.

1 is a system configuration diagram showing a configuration of a music search system 10 according to an embodiment of the present invention. 3 is a block diagram illustrating an example of a functional configuration of a lyrics data storage server 20. FIG. It is a figure which shows an example of the structure of the data stored in the lyric data storage part. 3 is a block diagram illustrating an example of a functional configuration of an audio data storage server 30. FIG. 4 is a diagram illustrating an example of a structure of data stored in a feature data storage unit 31. FIG. 3 is a block diagram illustrating an example of a functional configuration of an in-vehicle device 40. FIG. 4 is a diagram illustrating an example of a structure of data stored in a dictionary data storage unit 46. FIG. It is a figure which shows an example of the structure of the data stored in the music information storage part. It is a flowchart which shows an example of operation | movement of the music search system 10 at the time of music data acquisition. It is a flowchart which shows an example of operation | movement of the music search system 10 at the time of music data reproduction | regeneration. It is a hardware block diagram which shows an example of the computer 60 which implement | achieves the function of the lyrics data storage server 20, the audio | voice data storage server 30, or the vehicle equipment 40. It is a block diagram which shows the other example of a function structure of the vehicle equipment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Music search system, 11 ... Communication line, 12 ... Base station, 13 ... Vehicle, 14 ... Microphone, 15 ... Input device, 16 ... Speaker, 17 ... Antenna 18 Display device 19 Recording medium 20 Lyric data storage server 21 Lyric data storage unit 210 Song name 211 211 Singer name 212 .. Album name, 213... Lyric data, 22... Lyric data transmission unit, 30... Voice data storage server, 31... Feature data storage unit, 310. 32, feature data transmission unit, 40 ... in-vehicle device, 41 ... likelihood calculation unit, 42 ... word string identification unit, 43 ... music data extraction unit, 44 ... reproduction , 45... Feature quantity acquisition unit, 46... Dictionary data storage unit, 4 0 ... word, 461 ... feature, 47 ... lyric data acquisition unit, 48 ... music information storage unit, 480 ... song name, 481 ... singer name, 482 ... album name , 483 ... music data, 484 ... lyrics data, 49 ... music data acquisition unit, 50 ... second dictionary data, 51 ... first dictionary data, 52 ... command execution , 60 ... computer, 61 ... CPU, 62 ... RAM, 63 ... ROM, 64 ... HDD, 65 ... communication device, 66 ... input / output interface, 67 ... .Media interface, 68... Recording medium

Claims (7)

An in-vehicle device mounted on a vehicle,
For each piece of music data, music information storage means for storing lyrics data indicating the lyrics of the music,
Music data acquisition means for acquiring music data from a recording medium for storing music data and storing it in the music information storage means;
Lyrics data acquisition means for acquiring lyric data corresponding to the music data acquired by the music data acquisition means from outside and storing it in the music information storage means;
Voice recognition means for identifying a word string indicated by the voice from the voice spoken by the user;
Music data that refers to the music information storage means, identifies lyrics data that includes the same word string as the word string specified by the voice recognition means, and extracts music data associated with the specified lyrics data Extraction means;
A vehicle-mounted apparatus comprising: reproduction means for reproducing the music data extracted by the music data extraction means. The in-vehicle device according to claim 1,
The lyrics data acquisition means includes
An in-vehicle device, which is provided outside the in-vehicle device and acquires lyrics data via a communication line from a lyrics data storage server that stores lyrics data for each music piece. The in-vehicle device according to claim 2,
The lyrics data acquisition means includes
When the music data acquisition means acquires a new music and stores it in the music information storage means, the lyrics data corresponding to the music is acquired from the lyrics data storage server and associated with the music data. An in-vehicle device that is stored in the music information storage means. The in-vehicle device according to claim 2,
The lyrics data acquisition means includes
When the song data acquisition unit newly acquires a song and stores it in the song information storage unit, the user is inquired about whether or not the lyrics data corresponding to the song can be acquired, and the user is permitted to acquire the lyrics data. The vehicle-mounted device is obtained from the lyrics data storage server. The in-vehicle device according to any one of claims 1 to 4,
For each word, it further comprises a feature amount acquisition means for acquiring a feature amount of the designated word from an audio data storage server that stores information indicating the feature amount when the word is uttered,
The voice recognition means
Dictionary data storage means for storing each of a plurality of words together with information indicating a corresponding feature amount;
The voice received by the user is received through the microphone, the feature amount of the received voice is calculated, and the words stored in the dictionary data storage means are similar based on the calculated feature amount. Likelihood calculating means for calculating a likelihood indicating a higher value as the ratio is higher;
Word string specifying means for specifying a word string corresponding to speech received via a microphone based on the likelihood calculated by the likelihood calculating means;
The lyrics data acquisition means includes
When the lyrics data is newly acquired from the outside, if the word in the lyrics data is not stored in the dictionary data storage means, the feature quantity acquisition means is instructed to determine the feature quantity of the word. The in-vehicle apparatus characterized in that the information indicating the feature quantity acquired by the voice data storage server is stored in the dictionary data storage means together with the corresponding word. The in-vehicle device according to any one of claims 1 to 4,
For each word, it further comprises a feature amount acquisition means for acquiring a feature amount of the designated word from an audio data storage server that stores information indicating the feature amount when the word is uttered,
The voice recognition means
First dictionary data storage means used in speech recognition at the time of music search, storing each of a plurality of words together with information indicating a corresponding feature amount;
Second dictionary data storage means used in normal speech recognition other than music search, storing each of a plurality of words together with information indicating a corresponding feature amount;
The voice uttered by the user is received through the microphone, the feature quantity of the received voice is calculated, and the voice recognition at the time of music search is stored in the first dictionary data storage means based on the calculated feature quantity. Among words stored in the second dictionary data storage means in speech recognition other than the time of music search. A likelihood calculating means for calculating the likelihood at
Word string specifying means for specifying a word string corresponding to speech received via a microphone based on the likelihood calculated by the likelihood calculating means;
The lyrics data acquisition means includes
When lyric data is newly acquired from the outside, if the word in the lyric data is not stored in the first dictionary data storage means, the feature quantity acquisition means is instructed to The information indicating the feature amount is acquired from the voice data storage server, and the information indicating the feature amount acquired by the feature amount acquisition unit is stored in the first dictionary data storage unit together with the corresponding word. In-vehicle device. An in-vehicle device mounted on the vehicle;
A music search system that is provided outside the in-vehicle device and includes a lyrics data storage server that stores lyrics data for each song,
The in-vehicle device is
For each piece of music data, music information storage means for storing lyrics data indicating the lyrics of the music,
Music data acquisition means for acquiring music data from a recording medium for storing music data and storing it in the music information storage means;
Lyrics data acquisition means for acquiring lyrics data corresponding to the music data acquired by the music data acquisition means from the lyrics data storage server and storing it in the music information storage means;
Voice recognition means for identifying a word string indicated by the voice from the voice spoken by the user;
Music data that refers to the music information storage means, identifies lyrics data that includes the same word string as the word string identified by the voice recognition means, and extracts music data associated with the identified lyrics data Extraction means;
A music retrieval system comprising: reproduction means for reproducing the music data extracted by the music data extraction means.

Images