TW529018B - Terminal apparatus, guide voice reproducing method, and storage medium - Google Patents

Abstract

There is provided a terminal apparatus that eliminates the need for preparing information of pitch and intonation of a guide voice. Content data composed of performance data formed of a sequence of performance events, and voice symbol data formed of voice symbols indicative of respective syllables of lyrics attached to the performance data is distributed to the terminal apparatus. Musical tones are reproduced from the performance data. A guide voice is synthesized based on the voice symbol data. The performance data is pre-read to control the voice synthesis section, whereby characteristics of the synthesized guide voice are changed according to the performance data.

Description

529018 A7 B7 V. INTRODUCTION TO THE INVENTION (!) TECHNICAL FIELD The present invention relates to a terminal device and a guide capable of receiving content data to allow a user to play a backing vocal, and can be suitably applied to a backing device, a mobile phone, or a similar device Voice copy method and a storage medium for storing a program for performing a guided voice copy method. BACKGROUND ART Cellular telephone systems including Personal Digital Cellular telecommunication system (PDC) and Personal Handyphone System (PHS), which are known as digital cellular systems, occupy a relatively narrow bandwidth, so these systems transmit at a lower bit rate. data. This requires high-efficiency compression encoding to compress the state to transmit speech signals. Analysis-synthesis coding method is known as one of the methods of high-efficiency compression coding, which uses a speech synthesis model including a sound source model and a vocal tract model. Analysis-synthesis encoding methods include MPC (Multi-Pulse Excited LPC) and Code Excited LPC (Code Excited LPC) using codebooks to vector-quantize speech data. ). The CELP method has been adopted for practical use by specific types of digital honeycomb systems. Backing systems have traditionally been proposed to copy backing music from scattered backing materials to allow the user to play back backing, meaning singing the copied backing music. This vocal system is often referred to as the π communication karaoke system and includes a form that even distributes vocal data to the home. In this system, the sound of the required music is -4- this paper size applies to the Chinese National Standard (CNS) A4 specification (210X 297 mm) 529018 A7

Music data and guide lyrics data are displayed with visible lyrics prompts, and when required, the image data used as the background image is distributed with companion data. User's copy of the music sound song from the music data = At the same time watch the copy and display from the distributed guide lyrics data at: cover ^ guide lyrics (visible hint). However, the accompaniment system encountered a problem: when playing the accompaniment, it must be used while singing to watch the guide lyrics displayed on the screen, so the color is partially changed according to the performance of the first music. Therefore, if the user cannot see the display screen, it is often difficult for the user to use the backing vocal. This can happen, for example, when the user is driving a car, when there is no display, or the display is too small and the characters on the screen cannot be read. In order to solve the above problems, another type of communication accompaniment system has been published in Japanese tile patent publication (Japanese Laid-〇pen Patent

Recommended by Publicati〇n, Koka〇 No. 1 1-167392, in which a backing material composed of music data, background image data, lyrics display data as a guide lyrics, and vocal prompts attached thereto (v 〇Cal's lyrics data are transmitted together. A vocal device receives these data, and then copies the vocal music from the music data, and at the same time displays the guidance lyrics based on the lyrics display data according to the copy of the vocal music. On the screen showing the background image based on the background image data Furthermore, based on the accent (α (^ ηΪ) information, the intensity of the tone, and the pitch (pitch 'voice)) included in the vocal cue data, a synthesized sound is formed and then based on the The vocal prompts the reading timing information of the lyrics data to be output. Therefore, the user can listen to the speech synthesis by listening to it. 5- This paper size is suitable for B ® grid (210 X 297 mm). Explanation (,, ... reminds you to sing, don't need to look at the display screen. The word I sister 'corresponds to the song that the user sings Before the part, the vocal mention: Μ must be read or sung aloud, and the lyrics should be composed of lyrics data, so the synthesized sound:; == From the song user, the melody part of the singing melody /, you should use Γ The person feels that listening to the two causes the two to cause the vocalization to indicate that the lyrics data needs to include accent information, and the second: this y = the pitch (speech characteristics) of the synthesized sound and the reading sequence f. The analysis of similar parts of the melody and the song needs to be prepared for two benefits: use: for reasons, the system can be expected to "transmit the rate to a low bit rate, so the transmission capacity is also ΐ; :: Gen: therefore 'transmission has additional vocal prompting lyrics data :: Long time, it will increase the cost of using the system to call. Ϊ́ 二 2 After the user makes a request for the music piece, the backing vocal = will be distributed to the user 4 when the resource is lost for a long time, = After enduring the need for transmission, it is necessary to wait for a long time for the reproduction of music works, and it is known that the user may be interested in loosening the backing vocals. Eighth, the cellular phone must be equipped with speech synthesis equipment to synthesize sounds. Not only making things expensive, but also d. Shrinking puts restrictions because the speech synthesis device also needs space. In view of these circumstances, the present invention aims to provide a terminal device and a method for guiding speech reproduction to reduce the need for preparation Intonation information and 529018 A7 B7 V. Description of the invention (4) Guiding voice tones and a storage medium storing a program that executes the guidance voice reproduction method. In addition, a second object of the present invention is to provide a terminal device and a guidance voice reproduction method to Even with a low data transmission rate, the vocal data can be transmitted in a short time, and at the same time, a dedicated speech synthesis device is not needed for the copy of the guided voice and a storage medium for storing a program that executes the guided voice reproduction method. In order to achieve the above first object, according to the first aspect of the present invention, a terminal device is provided to disseminate content data composed of a series of performance data formed by a series of performance events, and a phase information attached to the performance data. Corresponds to the phonetic symbol data formed by the phonetic symbols indicated by lyrics syllables. According to the first aspect of the present invention, the terminal device includes a musical tone synthesis section including a reproduction of musical tones from performance data, a guide voice based on speech symbol data to synthesize a guide voice, and a pre-read performance data. The method controls the speech synthesis control section of the speech synthesis section to be characterized, so the characteristics of the synthesis guide speech are changed according to the performance data. According to the present terminal device, the speech synthesis section is controlled by pre-reading the performance data, so the characteristics of the synthesis guide speech are changed according to the performance data. Therefore, it is no longer necessary to prepare the sounding timing information of the guiding voice, as well as its pitch and intonation information. This eliminates the need to analyze the program of each musical composition and the accent information, pitch information, and pitch of the synthesized speech (the paper size of the speech applies to the Chinese National Standard (CNS) A4 specification (210 X 297 mm) 529018 A7 B7 V. Invention It is possible to explain (5) characteristics) information and read timing information. Furthermore, because the dispersed data does not need to include information to guide the pitch and pitch of the speech, it is possible to reduce the amount of data that must be distributed. In addition, because it is possible to control the timing of the sound of the guidance voice by pre-reading and analyzing performance data, the amount of data to be distributed can be further reduced. It is best that the performance data is MIDI format performance data, and voice symbol data is inserted into the performance data as a dedicated message. Preferably, the terminal device further includes an analysis section that analyzes performance data including vocal lines in the performance data, and a guide voice synthesized by the speech synthesis section based on the voice line and based on the analysis section change The pitch and pitch are used to analyze the results and control the speech synthesis section of the speech synthesis section. '' The speech synthesis control section is more suitable. The speech synthesis section controls the timing of the synthesis based on the analysis results of the synthesis section. Therefore, the guidance speech synthesized by the speech synthesis section corresponds to the utterance line corresponding thereto. Was sounded before. Further more suitably, the terminal device includes a voice database to store voice parameters, and the voice synthesis control section supplies voice synthesis from the voice parameters read from the voice database based on the voice symbol data and analysis results from the analysis section. Sections, whereby the guidance speech synthesized by the speech synthesis section has each syllable different from the phonetic symbol data-and the pitch and pitch can be changed according to the speech line. In order to achieve the above-mentioned second objective, according to the second aspect of the present invention, a terminal device is provided to disseminate performance data formed by a series of Φ performance events. -8- This paper size is applicable to the Chinese National Standard (CNS) A4 specification (210X 297 male) (Centi) 529018 A7 B7 5. Description of the invention (6) Content information of the phonetic symbol data formed by the syllable phonetic symbols corresponding to the instructions attached to the performance data. The terminal device according to the second aspect of the present invention includes a telephone function section including activation of a telephone voice, a music tone synthesis section that reproduces music tones from performance data, and a speech synthesis section that synthesizes guided speech based on speech symbol data To characterize and decode the phonetic voice data. According to the present terminal device, the guidance speech can be synthesized by using a speech synthesis section that decodes speech data as provided in a digital cellular phone handset, so the mobile phone handset does not need to have an additional speech synthesis section. Therefore, even if the mobile phone is set to output the guidance voice without the need for a new accommodation space, this makes it possible to reduce the size of the phone handset. Furthermore, the existing speech synthesis section can be shared for this purpose, and it is possible to prevent an increase in manufacturing costs. It is more appropriate that the terminal device further includes a speech synthesis control section that controls the speech synthesis section by pre-reading the performance data, so the characteristics of the synthesized guidance speech are changed according to the performance data. It is more suitable that the terminal device further includes an analysis section that analyzes the sound line performance data included in the performance data, and a speech synthesis control section that controls the speech synthesis section based on the analysis result of the analysis section to change the sound according to the utterance. The pitch and pitch of the guidance speech synthesized by the speech synthesis section. In order to achieve the first objective, according to the third aspect of the present invention, a method of copying the guidance speech is provided, which is distributed by using a terminal device by a series of performance events. The formed performance data and the instructions attached to the performance data-9- This paper size applies the Chinese National Standard (CNS) A4 specifications (210X 297 mm) 529018 A7 B7 V. The description of the invention (7) shows the corresponding lyrics syllables Content data of speech symbol data formed by speech symbols. The method is characterized by steps including copying musical tones from performance data, synthesizing guidance speech based on speech symbol data, and pre-reading the performance data such that the characteristics of the synthesized guidance speech are changed according to the performance data. In order to achieve the first object, according to a fourth aspect of the present invention, a storage medium is provided to store a program for causing a computer to execute a method for copying a guide voice. Brief description of the drawings. FIG. 1 is a diagram showing an example of a configuration of a cellular phone applied to a base station in accordance with the first embodiment of the present invention; FIG. 2 is a diagram showing a more detailed diagram The configuration settings of the voice compression and synthesis section of the cellular phone telephone function section in FIG. 1; FIG. 3 is a diagram showing the process of processing performance data and representing the functions of the processing section appearing in the telephone function section of FIG. Functional section diagram; Figure 4 shows the format of the backing material used by the cellular phone in Figure 1; Figure 5 is a conceptual rendering of the program for downloading backing material to the cellular phone in Figure 1; An example of the configuration settings of the accompaniment and hip devices implemented and applied in the distribution center; FIG. 7 is a diagram showing the configuration settings of the speech synthesis section of the speech synthesis section of the vocal device in FIG. 6 in more detail . Best Mode for Implementing the Invention -10- This paper size applies Chinese National Standard (CNS) A4 (210 X 297 mm) binding line 529018

Now, the invention will be described more fully with reference to the drawings showing specific embodiments shown herein. FIG. 1 is a diagram showing an example of a cellular phone configuration setting applied to a first specific embodiment of the present invention and a base station in parallel. In FIG. 1, reference numerals 1 and 1 represent the cellular phone according to the first embodiment of the present invention, and the reference numeral 2 represents a base station that manages a corresponding radio zone. Generally, digital honeycomb systems use a small-zone system in which the service area is divided into radio zones. The radio zones are each managed by a corresponding base station 2. When a call is made to a normal telephone, the cellular phone 1 and the base station 2 are connected to a parent switching station, and then connected to the Jintong telephone network through the switching port, as detailed below in this article. The cellular phone i is > provided for an antenna 10, which is usually of a retractable type and is connected to a transmitter ... grab ... a receiver / receiver section 11. The transmitter / receiver demodulates the signal received by the antenna ，, and modulates the signal to transmit to transmit the modulated signal to the antenna 10. The telephone function section 12 includes a processing device that can cause the cellular telephone 1 to function as a telephone to communicate with other telephones, and the speech compression-synthesis section 22 has a CELP encoder function and a CELP decoder function, both of which are adjustable for high Efficiency speech compression. The speech can be synthesized by supplying the speech parameters read from the database 24 to the speech compression-synthesis section 22 and using the speech compression-synthesis section 22 to synthesize the speech. In short, the language compression-synthesis section 22 can be triggered to function as a speech synthesis device. The database 24 stores voice parameters of “a” to “n” sounds and mimic sounds. During the call of DJ, the voice signal input through the microphone 2 1 every __________ -11- This paper is a standard GG home standard (CNS) A4 specification (2ι () χ297 public love) " —---- 、 Explanation of the invention (by speech compression-synthetic coding, cooking code) The code function receives high-efficiency compression-for transmission: 1. Modifier 1 Receiver 11 to modulate, and then pass the voice data _ U of antenna 10 through the antenna _Receiving efficiency compression coding function; to demodulate and then decode by phone and then compress 50% area & 22 to decode the original voice signal, including the output section of loudspeaker The output is 20. For example, during the telephone conversation *, the signal is transmitted or received through the transmitter / receiver σσ and 1 and the telephone function section 12.

, Storing a η horse as described later in this article temporary storage of scattered backing funds ^ memory tf. The accompaniment data is composed of the voice symbol data formed by the user f seeking _ material sound ㈣ 品 玩 一 所 ^ 的 演 4 资料 # The instructions attached to the performance data correspond to the three bone joint phonetic symbols. Further, the companion speech may include guide lyrics data showing lyrics on a display screen. As shown in Figure 1, the vocal data is written in MIDI format, and the phonological data of the lyrics is embedded with AMIDI data in a dedicated message. Therefore, the amount of backing material for a single music piece can be limited to a small size, which even H H /, a digital honeycomb system with a low 7L lean material transmission rate can transmit a piece of backing piece music piece in a short period of time. 'Data separating section (data separating section) 14 incorporates a MIDI decoder and resolves the MIDI data read from the storage device 13 to separate the same data into a musical instrument' data. Through the buffer memory (buffer memry, buff) 15 as a delay circuit, the separated performance data is supplied to a music sound synthesis area composed of a sequencer and a midi sound source. Section (mUSiCal t〇ne Synthesjs __ ~ 1 2-This paper size applies to China National Standard (CNS) A4 specifications (210X297 mm) binding 529018

sectl0n) 16. On the other hand, the separated speech symbol data is supplied to the telephone function section (telephone functión) together with the insipid material 12. In the telephone function section 12, the 'guide speech based on the speech symbol data is replaced by the speech The synthesis section 22 is compressed to synthesize and thereby output. The guidance voice provides an audio cue for the lyrics of the singer's backing song to help him sing in the place where the guidance ^ = appears on the display. The guidance voice is based on the synthesis of the tone from the paragraph 16 The process of dubbing the backing music tones is synthesized, and then output through the output paragraph 20. Therefore, each of the leading voices corresponding to a pre-measured lyrics phrase is synthesized and longer than the time that the lyrics phrase was sung Output early. In addition, after adding accents and tones, the guide is synthesized by a fast tempo according to the rhythm corresponding to the performance data. · To generate the output of the guide # 我 音 's timing and adding rhythm, The accent and tones are based on the same guiding voice, and the performance of the vocal lines included in the performance data is poor (the vocal part of the Section) is analyzed by the processing section of the telephone function section 12. For example, by analyzing the key changes, melody, and melody of the performance data of the vocal line, controlling how to guide the pitch change of the voice and by analyzing the response image It is the velocity line information (velocity) information and sounding duration (gate time) information of the musical score symbols such as slur and staccato, which controls the pitch and accent of the guiding voice Further, when + sing is used for duet, the tonal change of the vocal line performance data can be analyzed to determine whether a phrase is handled by a male or female part, and the guiding voice pitch of the phrase can be determined accordingly It is set so that the guidance voice sounds female or male. ____- 13- This paper size applies Chinese National Standard (CNS) A4 (210 X 297 mm) gutter 529018 A7

The speech symbol data supplied to the telephone function section 12 is transmitted to the database 24 'and the sound parameter "material library 24 is read out and supplied to the speech compression-synthesis section 22, so the sound represented by the speech symbol data is in syllables Based on synthesis by the speech compression-synthesis section 22. The speech parameters read out from the database 24 are produced through the control based on the analysis of the performance data of the speech line as described above, so the parameters reflect the melody, speed, and During the sound of the vocalization line, it has started the speech compression_synthesis section ^ The guidance speech ', so the pitch, accent, and pitch of the guidance speech are changed according to the vocalization line. As described above, the performance of the corresponding guidance speech The material part is pre-read and analyzed, and then the guidance voice is output before the musical tone is copied based on the performance profile part. In other words, the musical tone based on the performance profile is delayed until after the reproduction of the guide voice. This delay is buffered Memory memory 15). The buffer memory 15 is supplied with the infra shell material delayed by a predetermined period of time. The musical tones are synthesized into paragraphs 16 to duplicate the musical tones. As a result, the leading phrase synthesized by the compression / synthesis section 2 2 for the corresponding musical tones is copied through the musical tones synthesis section 16 to the corresponding musical tones via the output paragraph 20 The output is before. The music tone synthesis section contains a sequencer and a MIDI sound source. The music tones copied from the music tone synthesis section 16 are transferred to the effect section 17 where the effect 袂 ^ is added To music tones. The music tones with effects are synthesized by the synthesis section 18 together with the synthesis guide voice. Before being synthesized by the music tones, the guide voice has the effects added by the effects section 2 3. The music synthesized by the synthesis section 18 Tones and guiding voices are amplified by ____- 14- This paper size applies Chinese National Standard (CNS) A4 specifications (210 X 297 mm) 529018 A7 ^ __ B7 V. Description of the invention (12) Area & 19 The output section 20 is used for output. The effect sections 17, 23 depend on the number of speakers in the output section 20 for localization control. Furthermore, you can add Effects such as reverberation and chorus. Going one step further, the database only stores the sound parameters used to synthesize typical mechanical-sounding human voices, and thus the synthesized guidance voice can be used with an equalizer (equaHzer) ) To correct. Also, the volume of the guidance voice can be reduced according to the technique of operating the vocal singer.… Secondly, regarding the voice compression-synthesis paragraph 22 configuration of the telephone function section 12 of the cellular telephone 1 of FIG. 2 and the configuration together with the database The configuration of 24 is shown in more detail. The 7 ^ speech compression-synthesis section 22 shown in Figure 2 includes a typical CELP decoder for decoding highly efficient compression-encoded speech data. It should be noted that the compression-synthesis section 22 also includes a not-shown CELP encoder, which enables efficient compression coding of speech information. Now the basic principles of speech synthesis will be explained. The characteristics of speech can be described by the pitch L and the noise component of the source sound (noise component, which will be referred to as the source sound characteristic parameter), and when the sound passes through the speaker or singer's throat and Features of vocal-tract given to the voice at the mouth, and vocal features given to the voice when the voice passes through the speaker or the singer's lips (these features will be called horse vocal tract features Parameter "v〇caMract charactedstic parameters"). In short, the vocal chord model and vocal chord model (vocal tract m〇del) can be used to present the paper scale suitable for this paper] gg * standard (CNs) μ size tearing public Love) 529018

Now. The (: £ 1 ^ decoder in the language compression_synthesis section 22 in FIG. 2 decodes the compression-encoded speech data into the original sound by performing sound synthesis based on the speech synthesis mode described above. As shown in FIG. 2, Each data frame input to the speech compression-synthesis section 22 is separated into data parameters by the data processing section 30. The sound parameters are respectively index I, pitch L, and reflection coefficient r. The parameter indicated by the pitch L is supplied to the short term oscillating section 32, the parameter indicating the index I is supplied to the codebook 31, and the parameter indicating the response coefficient r is supplied to the throat approximation filter (throat). approximation filter) 34. The codebook 31 has the same contents as the encoder codebook used to encode the original sound. The contents of this codebook 31 are stored in the Read Only Memory (Read Only Memory) The short-term oscillating section 32 generates a decoded signal of the indicated sound at a pitch L based on the indicated pitch L parameter and transmits the generated encoded signal to a source waveform-reproducing section 33. It is supplied with Introduce the code vector data marked and bought from the codebook 3 11 to the source waveform copy section 33, and synthesize the data with the coded signal indicating the sound at pitch L to copy and synthesize the source waveform. From the source Waveform copy section 3 3 The output synthetic source waveform is similar to the waveform generated by human voice and chord vibration, and the filter coefficient is controlled by the indicator response system and the laryngeal approximation filter controlled by Qin parameters. 34 The synthetic source waveform is filtered into synthetic speech. The throat approximation filter 34, which duplicates the human throat and human mouth transfer functions, stores the response coefficient r supplied in advance from the data processing section 30 and sets the response coefficient _ -16-This paper size applies the Chinese National Standard (CNS) A4 specification (210X 297 mm) V. Description of the invention (14 = relative wave filter. The sum of the output from the throat approximation wave filter 34 is transmitted to the sound spectrum wave filter (ah ⑽ fmer) ^, not two outputs = Synthetic sound part (comp_ a) As the pressure of the speech signal formed by the M-rate compression coding method, · Boer-Yebeke was decoded and output by the speech compression and synthesis section "In addition, the data is separated The speech separated in section 14 is supplied to the sound database ㈣ ___, and the pitch parameters, the number of waveform selection tables, and the response coefficient parameters synthesized from the guided speech indicated by the poor speech symbols are transferred from the sound database. The output is supplied to the short-term oscillating section 32 with a pitch parameter of ⑽ not, and the short-term oscillating section ㈣ generates a decoded signal representing a sound having a tone. High L g and supplies it to the source waveform-copy section 33. The waveform selection parameters are supplied to a waveform database 41, and a waveform defining a sound pattern is read from the waveform database 41 and transmitted to a source waveform-copy section ^. The source waveform-copy section 33 synthesizes a decoded signal having a pitch Lg and a waveform defining a sound pattern to copy the synthesized source waveform. Output from the source waveform and copy section B < The synthesized source waveform is filtered by the throat approximation filter 34, and its tear wave coefficient is modified by the reaction coefficient modification database 42 from the response coefficient parameters supplied from the sound database 40 The read-out parameter indicating the response coefficient rg is controlled to guide the speech to be synthesized. The synthesized sound output from the throat approximation filter 34 is transmitted to the voice filter, the wave filter 3 whose unsynthesized pentatonic part which is unnatural to the human voice is eliminated, and is then outputted as a guide voice. In the above process, the database 24 is supplied with a control signal. This control signal controls the way of guiding the voice pitch Lg and the change of the pitch Lg as well as guiding the voice. 529018 A7 B7 V. Proof and Accent of the Invention. The control signal is embedded in the telephone function section 2 and the processing section represents the analysis result data made on the vocal performance data included in the performance data. By changing the oscillation frequency of the short-term oscillating section 32 and by controlling the sound parameter Lg using the control signal, it is possible to selectively synthesize and guide the voice into a male voice or a female voice. Furthermore, by changing the waveform data read from the waveform database 4 j, it is possible to change the type of the guidance voice sound. Furthermore, it is possible to change the pitch and accent of the guidance voice by changing the parameter indicating the response coefficient rg read from the response coefficient modification database 42. In the existing specific embodiment, since the control signal is generated based on the analysis of the sounding line performance data described above, the pitch, pitch, and accent of the guide voice can be changed according to the melody of the sounding line. As a result, the user can understand how the phrase and its tonality should be sung by listening to the other leading voices of the corresponding song. Further, the database 24 is supplied with and stores time information Dingime to indicate the sounding timing and the rhythm of the sounding guide ¥ 01065. In response to the time information D11, the predetermined waveform is read from the waveform database 41, and the parameters indicating the predetermined response coefficient ^ § are read from the reaction coefficient modification database 42. The time information Time is earlier than the leading voice rapture, and the lyrics are sung earlier. Furthermore, each tone of the guidance voice is controlled; the two lengths are controlled based on the time information Time, and the guidance voice is also output at this speed. The sound line performance data is analyzed by the processing section of the telephone function section 12 described above. This analysis is performed by executing an analysis program in the processing section. Figure 3 illustrates the ta ® ^^^ Cn57I4 ^ (21〇x 297 ^ «) ------- gutter line 529018 A7 B7 for the analysis process represented by functional blocks in the form of hardware components (16) Process. In FIG. 3, the MIDI data distributed as the vocal data is supplied to the storage device 13 from the transmitter / receiver section 11. The data separation section 14 reads out the MIDI data from the storage device 13 and interprets the MIDI data using the MIDI decoder function to separate the data into performance data and voice symbol data. As shown in FIG. 4, the MIDI data has voice symbol data and is inserted here as an external message represented by a sandwiched portion between the state bytes M F0 ″ and “F7” as shown in the figure. The external message is provided in a series For each phrase, a phonetic symbol indicating the guidance voice is formed and can include? 丨 Timing data for which the guide phrase should be spoken. _ The voice symbol data separated by the data separation section 14 is supplied to the guidance voice pitch / In the tempo-determining section 36, the pitch and tempo of the guidance voice are determined here. In this case, it is possible to mark the pitch to the user in order to select the guidance voice as Female or male voice. In addition, pitch information and rhythm data as well as pitch data and accent data are supplied from the vocal line-analyzing section 38 to the guidance voice through a switch (SW) Pitch / rhythm determination section 36. The guidance voice pitch / rhythm determination section 36 outputs control signals based on various types of information received with the voice symbol data, because This causes the guidance voice to be formed, which has the indicated pitch and is dependent on the vocal line. The control, Sister South L controls the pitch, speed (rhythm), intonation, and stress of the guidance voice. It should be noted When the switch SW is in the OFF state and the pitch is not marked, the guidance voice is formed to have a predetermined pitch. If the guidance-1 9-This paper size applies the Chinese National Standard (CNS) A4 specification (210X (297 mm) 529018 5. Description of the invention (17) The speech sound is unnatural. When it is output according to the beat of the sound line, only the parent switch SW is turned off. In this case, the monophonic guidance speech is turned off. The performance data separated by the data separation section 14 is supplied to the vocal line_analysis section 38, and the vocal line performance data included in the performance data is analyzed here. The child separation performance data is delayed in the buffered memory 15 After that, it is also transmitted to the Qingyue tone synthesis section 16. As a result, the musical note based on the performance data is copied after being delayed and the voice is read, and at the same time it is read and analyzed by the phonetic-analysis section. The sound beam_analysis section is called speed information and envelope information, which reflects the change of sound lines and musical tonality (melody), such as portamento and staccato. In addition, durat information and etime information It is also analyzed. The melody is poor from the analysis, and the vocal ray-analysis section 3 ancient guide speech articulates U ... 乂 " Control information is supplied to 5 丨 translation π, θ, and / or immediate decision section 36. From Speed information and message information analysis: The obtained accent control information and pitch control information are also supplied to guide the tongue five-bone elevation / rhythm determination section 36. Further, the guidance voice pitch / rhythm ^ section 36 is supplied with the guidance ancestral timing information and the guidance voice rhythm information obtained from the period information and E time information analysis. "* It should be noted that when the musical piece is duet, the vocal line-analysis section 38 analyzes whether the vocal line has a male voice or a female voice, and transmits the pitch information of the analysis result to the pitch / rhythm of the oblique voice. · Decision ^ Therefore, the pitch of the guiding voice and the length of each syllable are controlled according to the melody of the _ 'line. And, in the case of duet, the output is :; This paper scale applies the Chinese national standard Qi Qing 8 4 specifications ( 2 Staple-20- 1〇X 297 ^ ¾)

The synthesized guidance voice is synthesized by the male voice; the second is that the voice is okay, and the timing of outputting the male voice at the same time depends on the second ear from the axe. Each pilot voice output sounds timing information. It should be provided with the guidance voice signal provided by ^ 38 from the pitch of the guidance voice;; play :::: ring: sequence information and rhythm funds. Furthermore, when the reading information is transferred to level 1, the guide word: Two! The material contains the timing sequence of the guide sound. The earth sounds in this time series information. Then the control signal (lnterP01at〇r) 37 from the guide ★ Wu Li Li Gu, ^ through the interpolation% decision section 36 is output to the asset library 24. Xing Li corresponds to the melody of the vocal line. 贝贝 村 厍 24 Prevents the voice of the guide voice from changing unnaturally according to the phase. -...: The interpolator dynamically sets the rate of change of the guide according to the melody of the vocal line. This causes the material speech to be output in a normal voice. How?> Wang Yi realized that the buffer memory provides a sequence of ^ copied by synchronized voice lines to guide the day sequence of voice sounds, and the aforementioned sequence of guidance voice sounds is caused by buffer memory 15 Considering the delay time, the cellular phone to which the terminal device according to the present invention is applied can download the vocal data from the outside. Fig. 5 is a conceptual description showing how the backing material is downloaded to the cellular phone 1a and the cellular phone 1b, each of which is constructed similarly to the cellular phone 1 of Fig. 1 used by the terminal device ^ soul of the first embodiment of the present invention. The hive system of the hive is a small area system in which each service area is divided into several radio areas. The radio zones are each managed by the base stations provided here. When a cellular phone, such as a base station, makes a call to the general paper size, the Chinese National Standard (CNS) A4 specification (210X297 mm) is bound 529018. 5. Description of the invention (19 = machine, the cellular phone is managed by the radio zone to which the cellular phone belongs. Γ 占 is connected to the mobile parent station (m〇blle eXChange Station), and then to the general telephone network. Therefore, through the radio network, the honeycomb can be connected to the base station that manages the corresponding radio area, so that the honeycomb ^: Yes Make a call to another phone. Furthermore, when the honeycomb phone speaks to: 2; other honeycomb phones in other radio areas, the transmission-side cell phone j passes through the base station that manages the transmission-side cell phone radio, connects to the mobile station, and then switches to The mobile switching station is connected to the base station of the relevant cellular telephone on the receiving side. Figure 5 illustrates the example of the cellular system described above, where the cellular telephone "belongs to a radio area managed by one 2c of the base stations 2a to 2d, while the cellular electrical activity lb belongs to the base station -2a radio area managed. Cellular phone h is connected to the base by radio network Station 2c. ㈣ Nest electric fela is used for call or location note up-signal received and processed by the base station 2c. Although the cellular phone ib is managed by the base station 2a, the relationship between them is similar to that between the cellular phone h and the base station 2c The base station M2d manages different radio areas respectively, but the radio areas can have perimeters that overlap each other. The base stations "to 2d are connected to the mobile switching station 3 through a multiplex line. And several mobile switching stations 3 are condensed to connect to Jin The gate exchange station (0 ###) of the telephone exchange station 5a 4. Several gate exchange stations.-4 are interconnected through a relay transmission line plus a tfa_iSsion line. Ordinary spring language exchange station Maru ... is located in the corresponding local area (10cal areas), and is also interconnected through a relay transmission line. Each ordinary telephone exchange station, 5b, brother ... has multiple ordinary telephones connected here, and for example, distribution center 6 is connected to ordinary telephones-22- This paper standard applies to China National Standard (CNS) A4 specifications (21〇χ297 (Mm) 529018 A7 V. Description of the invention (20 exchange station 5b. 4! In Γ, multiple vocal material items are stored and new songs can be edited via :, ϋ ::, this. According to this specific embodiment, it is possible to replace The backing data is downloaded from the distribution center 6 connected to the ordinary telephone network to the honeycomb phone la, = when the honeycomb phone 1a is downloading the backing data, it transmits the electric cleaning number of the center V. As a result, the hive Nanhe % Meh is connected to the distribution through the path of the base station 2c, mobile exchange "A parent station 4, ordinary telephone exchange station 5a and ordinary telephone exchange station 5b. Then, by operating a numeric key pad, The following guidance of the guide is displayed on the screen of the hive phone h. When the screen is similar to jog dial or the hive phone 4, the user can request a vocal data item with the required song head up and download it. Similar things. In this case, the vocal vocal contains the voice symbol data of the guide voice. Similarly, the honeycomb phone lb can request and download the vocal data items with the desired song heading up. The distribution center 6 can be connected to the Internet, This allows the backing material to be downloaded from the distribution center 6 via the Internet. When the backing is played using the cellular telephone 1 shown in FIG. 1, the rest voice input from the microphone 21 is also output from the output section 20. In this case, Hive Qi 1 is enough to perform speakerphone talk ea⑴, and thus when the vocal is played in the speakerphone state, the sound output from the output section 20 can be input through the microphone 2J: cool input, so _ roar ( howling). In order to overcome this problem, when the hive phone 1 is set up to be able to talk hands-free, an echo cancellation circuit (ech〇canceUer circuit) is provided to prevent howling. Then the FM modulator can be ____ -23- This paper size applies to China National Standard (CNS) A4 (21 × 297mm) binding 529018 A7

Because the radio is weak, the device is installed in the room to transmit the output signal by the input section 20, and the FM receiver can receive the snoring sounds in the two compartments (Vehkle (called 咖 nt)). And in this case, sound. Therefore, a cancellation circuit is provided to prevent howling and 'when the vocal is played, the transmitter section of the cellular telephone 1 is used for additional requests for transmission of requests, and thus for supplying power to the transmission standby area & The power supply can be maintained in the OFF state, except when a demand is made, thereby extending the service life of the battery. FIG. 6 shows a configuration example of a backing device of a terminal device applied to a second embodiment of the present invention, and a distribution center. This embodiment is basically different from the first embodiment in terms of f == and display function. More specifically, the second embodiment differs from the first embodiment in that the modem 111 and the control section 112 correspond to the transmitter / receiver function section 11 and the telephone function section of the first embodiment, respectively. The configuration of 12 is different, and the display section 126 is provided separately, but the other components of this embodiment and the components corresponding to the first embodiment have the same functions, and thus other components use the same reference numbers. Alas, and a detailed description is omitted here. In FIG. 6, reference numeral 100 indicates that the backing device of the terminal device applied according to the second specific embodiment and the back-up device 100 can download backing data from the distribution center 6. The backing device 100 and the distribution center 6 are connected to each other through a communication line. For example, a communication line may be formed by a telephone line. The backing device 1 〇〇 includes the modem 111. The desired backing data items can be distributed from the distribution center 6 _ -24- This paper size applies the Chinese National Standard (CNS) A4 specification (210X 297 mm) binding line 529018 A7 ------_____ B7_ V. Description of the Invention (22) ''-Download. The modem in demodulates the received signal and modulates the signal to be transmitted and then transmits the modulated signal to the communication line. The control section 112 includes a display control section 125 and a speech synthesis section 122. The control area 112 controls the overall operation of the backing device 110. When the speech is synthesized by the control area end two, the speech parameters read from the database 24 can be supplied to the speech synthesis section 122 'and the speech synthesis section 122 can synthesize speech based on the speech parameters. The data 24 stores the voice parameters from “a” to, “n” sounds and imitati sounds. Similar to the storage device 13 in the first embodiment, the storage device is a memory for storing the dispersed backing vocals. The accompaniment data is composed of the performance data formed by the user's request, the music product-it string event, and the language symbol data is composed of the bamboo syllables attached to the performance data indicating the respective syllables of the lyrics, and The guide lyrics display data for displaying lyrics is on the device 126. The guide lyrics display data is supplied from the modem lu to the control section 112. When the performance data is played, the guide lyrics display data is transmitted one by one from the control section 112 Go to the display 126, # This corresponds to the guide lyrics of each paragraph are sequentially displayed on the display 126. At the same time, the background image data is read from the large-capacity storage device and is not displayed on the map. Second, it is related to the guide lyrics. It is displayed on display 126. In addition to the guide lyrics display data, the accompaniment data is in MIDI format. As shown in Figure 4 by Lin Li, additional information is inserted into the MIDI data. Therefore, the amount of data for a single piece of music in the accompaniment can be reduced to a small amount, making it possible for a piece of accompaniment to be transmitted in a short time. -25- This paper size applies to the Chinese national standard (CNi ^ A4 specification (210X 297 mm) 529018 A7 B7 V. Description of the invention (23) In this specific embodiment, the speech separated by the data separation section 14 The symbol data is supplied to the control section 112 together with the performance data. The control section 112 synthesizes the guidance speech based on the speech symbol data and transmits the synthesized guidance speech from the speech synthesis section 122. The guidance speech is provided as a vocal prompt Help the user to sing without looking at the guide lyrics displayed on the display section 126. According to the music tone synthesis section 16, the music tones of the backing music are copied, the guide voice is synthesized, and then through the output section 20 Output. When the accompaniment is played by the accompaniment device 100, the singing voice input through the microphone 21 is also output from the output section 20. Then, with reference to FIG. 7, the root In the second specific embodiment, the configuration settings of the speech synthesis section 122 of the control section 112 and the database 24 of the backing device 100 are shown in more detail here. The speech synthesis section 122 shown in FIG. 7 is different from the first The speech compression-synthesis section 22 of the cellular phone 1 to which the terminal device of the specific embodiment is applied, because no encoder is provided here. The other parts of the configuration are similar to the speech compression-synthesis of the cellular phone 1 in configuration settings. The other parts of the section 22, and therefore detailed descriptions are omitted herein. Further, similar to the first specific embodiment, the performance data of the sound line is analyzed by the processing section of the control section 112. This analysis process is performed by the processing area of the control section 112 and the execution of the analysis program. The flow of this process is similar to that performed by the analysis process performed by the hive phone 1 of the first specific embodiment, as shown in FIG. 3, and therefore a detailed description is omitted herein. -26- This paper size applies the Chinese National Standard (CNS) A4 specification (210 X 297 mm) 529018 A7 B7 V. Description of the invention (24) Now, a description will be given of the backing device 100 how to download the backing data. The backing device 100 accesses the distribution center 6 through the modem 1. As a result, the backing device 100 is connected to the distribution center 6. Then, when the operation input device is not displayed, and when the following follow-up instructions are displayed on the display 126, the user can request a backing data item with the desired song head up and download the same track. In this case, the accompaniment data includes the voice symbol data of the guide voice and the guide lyrics display data attached here. It should be noted that the distribution center 6 can be connected to the Internet to thereby allow backing material to be downloaded from the distribution center 6 through the Internet. -Needless to say, the purpose of the present invention can be accomplished by installing software code to achieve the terminal device function of the above specific embodiment. From the storage medium of the electronic device in which the code is recorded, such as a vocal device , Cellular telephone, or personal computer (PC), and the computer (or CPU) that triggers the electronic device to execute the program. In this case, the program code installed in the electronic device using a storage medium reaches a new function of the invention, and the storage medium storing the program code constitutes the present invention. The storage medium for recording the code may be, for example, a floppy disk, a hard disk, an optical memory disk, a magneto-opticai ^ 4isk, a CD-ROM, CD-R (CD-Recordable, magnetic tape, nonvolatile memory card, or ROM. In addition, the code can be sent from the server computer through the communication network -27- This paper size is in accordance with China National Standard (CNS) A4 (210X 297 mm) 529018 A7 — B7 " "-" 丨 丨 " 丨 _ 5. The invention description (25) is not required It goes without saying that the present invention includes a case in which the illustrated embodiment functions are not only performed by executing a code read by the computer, but also cause an operating system (OS) operating on the computer to perform operations according to the code. Instructions, execute part or all of the operation. Furthermore, the present invention also includes a case in which a code read from a storage medium is written on an extension function board inserted in a backing device or a pc or connected to Extended function The memory provided in the unit, and the CPU or similar object integrated in the extended function board or the extended function unit perform part or all of the operations based on the instructions of the code to complete the functions of the illustrated embodiment. Industry Industrial applicability As described above, the terminal device according to the present invention can be applied to a vocal device with a communication function and a mobile phone, so a cellular phone or a car phone can have a vocal function. Furthermore, it is It is possible to apply the terminal device of the present invention to an electronic device having a backing function by providing a communication function for connecting a modem, a cellular phone, or the like to the device. Figure 1 1 Cellular Phone 2 Base Station (BASE STATION) 10 antenna (ANTENNA) 11 transmitter / receiver section, f5 & RANSMITTER / RECEIVER SECTION) 12 telephone function section (TELEPHONE FUNCTION SECTION) 13 storage device (STORAGE MEANS) ____- 28- This paper size is applicable to China Standards (CNs) A4 specifications (210X297 mm) 529018 A7 B7 V. Description of invention (26) MIDE MIDE DATA 14 DATA SEPARATING SECTION VOICE SYMBOL PERFORMANCE DATA 15 BUFFER MEMORY (BUFF) 16 MUSICAL TONE SYNTHESIS SECTION ) 17 EFFECT SECTION 19 AMPLIFIER SECTION 20 OUTPUT SECTIQN 2 1 MICROPHONE · 22 VOICE COMPRESSION- SYNTHESIS SECTION 23 Effect Section (EFFECT SECTION) 24 database (DATABASE) Figure 2 22 VOICE COMPRESSION-SYNTHESIS SECTION Compression VOICE DATA 24 DATABASE 3 0 DATABASE section PR © Name ESSING SECTION) 3 1 CODE BOOK 32 Short-term Oscillation Section (SHORT TERM OSCILLATING SECTION) -29- Binding Line This paper applies Chinese National Standard (CNS) A4 specifications (210X297 mm) 529018 A7 B7 V. Description of the invention (27) 3 3 SOURCE WAVEFORM-REPRODUCING SECTION 3 4 THROAT APPROXIMATION FILTER 3 5 SPECTRAL FILTER OUTPUT 4 0 VOICE DATABASE CONTROL SIGNAL VOICE SYMBOL 4 1 Waveform database ( WAVEFORM DATABASE) 4 2 REFLECTION COEFFICIENT MODIFYING DATABASE Figure 3 MIDI DATA 12 TELEPHONE FUNCTION SECTION 13 STORAGE MEANS VOICE SYMBOL performance data ( PERFORMANCE DATA) 15 BUFFER to TO MUSICAL TONE SYNTHESIS SECTION 3 6 guide voice pitch / tempo-determining section Designate PITCH ) ___ -30- This paper size is applicable to Chinese National Standard (CNS) A4 (210x 297 mm) 529018 A7 B7 V. Description of the invention (28) 3 7 Interpolator Control signal + voice symbol (VOICE SYMBOL) to the database (TO DATABASE) 3 8 vocal line-analyzin g section) Figure 4 MIDI DATA Sequence of voice symbols used for voice guidance (SEQUNECE OF VOICE SYMBOLS FOR VOICE GUIDE) Extra message (EXCLUSIVE MESSAGE) Figure 5 · 2a BAS'E STATION 2b BASE STATION) 2c base station (BASE STATION) 2d base station (BASE STATION) 3 mobile exchange station (MOBILE EXCHANGE STATION) 4 gate exchange station (GATE EXCHANGE STATION)

5a ORDINARY TELEPHONE EXCHANGE STATION

5b ORDINARY TELEPHONE EXCHANGE STATION

5c Ordinary Telephone Exchange Station (ORDINARY TELEPHONE EXCHANGE STATION) 6 Distribution Center (DISTRIBUTION CENTER) ___- 31 _-__ This paper size applies to China National Standard (CNS) A4 specifications (210X 297 mm) 529018 A7 B7 V. Description of the invention (29 ) Figure 6 6 DISTRIBUTION CENTER

MODEM 13 STORAGE MEANS MIDI DATA 1 4 DATA SEPARATING SECTION VOICE SYMBOL PERFORMANCE DATA 15 BUFFER MEMORY (BUFF)) 16 MUSICAL TONE SYNTHESIS SECTION 17 EFFECT SECTION 1 9 AMPLIFIER SECTION 2 0 OUTPUT SECTION 2 1 MICROPHONE 2 3 EFFECT SECTION 2 4 DATABASE 100 KARAOKE APPARATUS 112 CONTRL SECTION 122 VOICE SYNTHESIS SECTION 125 DISPLAY ^ OONTROL SECTION 126 DISPLAY Figure 7 122 VOICE SYNTHESIS SECTION _ -32- This paper size is applicable to China National Standard (CNS) A4 (210X 297 mm) Gutter 529018 A7 B7 V. Description of the invention (30 ) 2 4 DATABASE

3 2 SHORT TERM OSCILLATING SECTION 3 3 SOURCE WAVEFORM-REPRODUCING SECTION 34 THROAT APPROXIMATION FILTER 3 5 SPECTRAL FILTER OUTPUT 40 VOICE DATABASE CONTROL SIGNAL VOICE SYMBOL 4 1 WAVEFORM DATABASE 4 2 REFLECTION COEFFICIENT MODIFYING DATABASE -33 · This paper is for China National Standard (CNS) A4 specification (210 × 297 mm)

Claims (1)

6. Scope of Patent Application 1. A terminal device, which is distributed with content data composed of a series of performance events formed by a series of performance events, and voice symbol data formed by voice symbols indicating individual lyrics syllables attached to the performance data The terminal device includes: a musical tone synthesizing section that copies musical tones from the performance data; a voice synthesizing section that synthesizes guided speech based on the voice symbol data; a voice synthesizing control section that uses pre-reading The performance data controls the speech synthesis section in such a way that the characteristics of the synthesis-guided speech change according to the performance data. 2. If the patent application is for the terminal device of item 1, wherein the performance data is MIDI format performance data, it has voice symbol data that will be inserted into the performance data with additional information. 3. For example, the terminal device of the first patent application scope further includes an analysis section that analyzes performance data including sound lines included in the performance data, wherein the speech synthesis control section controls the speech synthesis area based on the analysis result of the analysis section. Segment to change the pitch and pitch of the guide speech to be synthesized by the speech synthesis section according to the utterance line. 4. If the terminal device of the third scope of the patent application, the speech synthesis control section controls the synthesis timing of the speech synthesis section based on the analysis and results of the synthesis section, so the speech synthesis section is synthesized by the speech synthesis section. The guidance speech is made before the corresponding sound line. 5. If the terminal device in the scope of patent application No. 4 includes stored voice parameters -34- This paper size applies Chinese National Standard (CNS) A4 specifications (210 X 297 mm) AB c D 529018 A database, wherein the speech synthesis control section reads the speech parameters from the speech database based on the speech symbol data and supplies it to the speech synthesis section, so that the guidance speech synthesized by the speech synthesis section has a correspondence with the speech symbol data Each syllable of the sound, and the pitch and pitch here can be changed according to the sound line. 6. A terminal device that is disseminated with content data composed of performance data formed by a series of performance events and voice symbol data formed by speech symbols attached to the performance data indicating individual lyrics syllables, the terminal device Contains: ~ Install a phone function section to initiate telephone speech; a music tone synthesis section to copy music tones from performance data; and a speech synthesis section to synthesize guidance based on the speech symbol data Voice and decode voice data for phone calls. Line 7. For example, the terminal device under the scope of patent application No. 6 further includes a speech synthesis control section, which controls the speech synthesis section by pre-reading the performance data. In this way, the characteristics of the synthesized guidance speech are based on the Performance data to change. 8. As for the terminal device under the scope of application for patent item 6, wherein the performance data is MIDI format performance data, it has voice symbol data such as additional information that will be inserted into the performance data. 9. If the terminal device under the scope of patent application No. 6 further includes an analysis section for analyzing performance data including sound lines included in the performance data, the term -35- This paper standard applies to China National Standard (CNS) A4 specifications ( (210X 297mm) AB c D 529018 VI. Patent application range The sound synthesis control section controls the speech synthesis section based on the analysis result of the analysis section, so as to change the guidance speech to be synthesized by the speech synthesis section according to the sound line Pitch and pitch. 10. For the terminal device with the scope of patent application No. 9, wherein the speech synthesis control section controls the synthesis timing of the speech synthesis section based on the analysis result of the synthesis section, so the guidance speech synthesized by the speech synthesis section is in The vocal line corresponding to this sounds before. 11. The terminal device according to item 9 of the scope of patent application, includes a voice database storing voice parameters, and the voice synthesis control section reads voice parameters from the voice database based on the voice symbol data and supplies the voice synthesis section to the voice synthesis section. Therefore, the guide speech synthesized by the speech synthesis section has each syllable that matches the material of the speech symbol, and the pitch and pitch here can be changed according to the utterance line. 12. A method of guiding speech playback, which is a speech symbol formed by using a piece of content data composed of performance data formed by a series of performance events, and a speech symbol indicating individual lyrics syllables attached to the performance data For the terminal device of data, the method includes the following steps: copying the musical tones from the performance data; synthesizing the guidance speech based on the voice symbol data; and pre-reading the performance data to save money ^ and the characteristics of the synthesized guidance speech are based on the Performance data to change. 13. —A storage medium whose storage program is used to cause a computer to use a composition composed of a series of performance events formed by distributing performance data. -36- This paper size applies to China National Standard (CNS) A4 specifications (210X 297 mm) gutter 8 8 8 8 AB c D 529018 6. The patent application scope includes the terminal device of the phonetic symbol data formed by the phonetic symbol data of the individual lyrics syllables attached to the performance data, and executes the copy guide voice The method includes: a module for copying musical tones to copy the guidance speech from the performance data; a guidance speech synthesis module for synthesizing the guidance speech based on the speech symbol data; and a guidance speech change The module is used to pre-read the performance data, and in this way, the characteristics of the synthesized guidance voice are changed according to the performance data. -37- This paper size applies to China National Standard (CNS) A4 (210X 297mm)