JP5953743B2 - Speech synthesis apparatus and program - Google Patents

Description

  The present invention relates to a speech synthesizer and a program.

  When a character string such as lyrics and pitches and lengths of a plurality of notes (hereinafter referred to as “note strings”) are input as data, voice synthesis is performed that synthesizes the sound of a singing sound according to the character string and sound. An apparatus is known (for example, Patent Document 1). Moreover, in such an apparatus, a technique has been proposed in which a voice feature is extracted from a singing voice of a singer and a synthesized voice is edited using the extracted feature (for example, Patent Document 2).

JP 2006-259768 A JP 2009-217141 A

By the way, in the speech synthesizer as described above, in order to confirm whether the synthesized speech has been edited as intended, the user needs to reproduce the synthesized speech after inputting the singing speech. It was. In this case, in order to generate synthesized speech using features extracted from speech, it is necessary to perform various kinds of complicated processing. However, such processing has a large processing load and requires time. For this reason, the user needs to wait until the synthesized speech is reproduced, and the synthesized speech editing process may not be performed smoothly.
The present invention has been made in view of the above-described background, and in a device that generates synthesized speech using features (attributes) extracted from speech, it is possible to easily check the generated synthesized speech. The purpose is to provide.

To solve the problems described above, the present invention includes generating a receiving unit for receiving the attribute data indicating an attribute containing the pitch and volume of the audio data, the synthesized speech based on the previous SL attribute data receiving unit has received And a sound signal generation unit that generates a sound signal having periodicity based on the pitch and volume indicated by the attribute data received by the reception unit before the speech synthesis unit generates synthesized speech. The speech synthesizer further includes a sound signal generation unit that generates a sound signal having periodicity by a process of a shorter time than a process of generating the synthesized speech .

  In a preferred aspect of the present invention, a second receiving unit that receives lyric data indicating lyrics and score data associated with the lyrics, a pitch indicated by attribute data received by the receiving unit, and the second An association unit that associates the score data received by the receiving unit with each other and associates the lyrics data with the pitch data representing the pitch based on the association result, and the speech synthesis unit May generate synthesized speech based on the attribute data received by the receiving unit and the lyric data and pitch data associated by the associating unit.

In a further preferred aspect of the present invention, the audio data may be analyzed for attributes including pitch and volume , and an audio analysis unit may be provided that supplies attribute data indicating an analysis result to the receiving unit.

Further, the present invention is a computer to receive function to receive attribute data indicating an attribute containing the pitch and volume of the audio data, and voice synthesis function that generates synthesized speech based on the previous SL received attribute data, wherein A sound signal generating function for generating a sound signal having periodicity based on a pitch and a volume indicated by the received attribute data before the voice synthesizing function generates a synthesized voice, wherein the voice synthesizing function generates a synthesized voice; Provided is a program for realizing a sound signal generation function for generating a sound signal having periodicity by processing in a shorter time than processing to be generated .

  ADVANTAGE OF THE INVENTION According to this invention, in the apparatus which produces | generates a synthetic | combination voice using the characteristic (attribute) extracted from an audio | voice, confirmation of the synthetic | combination voice produced | generated can be performed easily.

Block diagram showing an example of a hardware configuration of a speech synthesizer The figure which shows an example of the content of song score data The figure which shows an example of the screen displayed on a display part Block diagram showing an example of the functional configuration of a speech synthesizer Flow chart showing the flow of processing performed by the control unit

<Embodiment>
<Configuration>
FIG. 1 is a block diagram illustrating an example of a hardware configuration of a speech synthesizer 100 according to an embodiment of the present invention. The speech synthesizer 100 is a device that synthesizes speech based on musical score data including a character string and a phoneme sequence and outputs the synthesized speech. The voice synthesizer 100 includes a control unit 10, a storage unit 20, an operation unit 30, a display unit 40, a voice processing unit 60, a microphone 61, and a speaker 62, and these units are connected via a bus 70. The control unit 10 includes a CPU (Central Processing Unit), a RAM (Random Access Memory), a ROM (Read Only Memory), and the like. In the control unit 10, the CPU controls each unit of the speech synthesizer 100 by reading a computer program stored in the ROM or the storage unit 20, loading it into the RAM, and executing it. The operation unit 30 includes various operators and outputs an operation signal representing the content of an operation performed by the user to the control unit 10. The display unit 40 includes a liquid crystal panel, for example, and displays various images under the control of the control unit 10.

  The microphone 61 outputs an analog audio signal representing the collected audio to the audio processing unit 60. The audio processing unit 60 includes an A / D (Analog / Digital) converter, converts the analog audio signal output from the microphone 61 into digital audio data, and outputs the digital audio data to the control unit 10. get. The audio processing unit 60 includes a D / A (Digital / Analog) converter, converts digital audio data received from the control unit 10 into an analog audio signal, and outputs the analog audio signal to the speaker 62. The speaker 62 emits a sound based on the analog audio signal received from the audio processing unit 60. In this embodiment, the case where the microphone 61 and the speaker 62 are included in the speech synthesizer 100 will be described. However, the speech processing unit 60 is provided with an input terminal and an output terminal, and the input terminal is connected via an audio cable. An external microphone may be connected, and similarly, an external speaker may be connected to the output terminal via an audio cable. In this embodiment, the audio signal output from the microphone 61 to the speaker 62 is an analog audio signal. However, digital audio data may be input / output. In such a case, the audio processing unit 60 does not need to perform A / D conversion or D / A conversion. The same applies to the operation unit 30 and the display unit 40, and an external output terminal may be provided to connect an external monitor.

  The storage unit 20 is a storage unit for storing various data, and is, for example, an HDD or a nonvolatile memory. The storage unit 20 includes a Timbre database 21, a phonological template database 22, a singing score data storage area 23, a singing voice data storage area 24, and an analysis result data storage area 25, as illustrated. The Timbre database 21 is a database in which voice parameters having different phoneme names and pitches are collected. This database is a database that the control unit 10 refers to when performing speech synthesis from singing score data. The voice parameters can be classified into, for example, an envelope of an excitation waveform spectrum, an excitation resonance, a formant, and a difference spectrum. These four speech parameters are obtained by decomposing the spectral envelope (original spectrum) of the harmonic component obtained by analyzing actual human speech or the like (original speech). A voice at a certain time can be expressed by a voice parameter (a set of excitation spectrum, excitation resonance, formant, and difference spectrum), and a voice parameter expressing the same voice is different if the pitch is different. The Timbre database 21 has phoneme names and pitches as indexes. Therefore, the control unit 10 can read out the voice parameters at a certain time t using the data belonging to the phonological track and the pitch track of the singing score data as keys.

  The phoneme template database 22 stores phoneme template data. The phoneme template data is data applied to the transition interval between phonemes and phonemes in the singing score data. When a human utters two phonemes in succession, it changes slowly, not suddenly. For example, if the vowel “e” is pronounced continuously without placing a break after the vowel “a”, “a” is pronounced first, and the pronunciation located between “a” and “e” After that, it changes to “E”. Therefore, in order to perform singing synthesis so that the phoneme combination part becomes natural, it is preferable to have some form of speech information of the excitement part for phoneme combinations that can be combined in a certain language. Considering this, preparing the voice data and the amount of pitch fluctuation in the phonological transition section as template data, and applying this template data to the phonological transition section in the singing score data, it is closer to the actual singing Realize speech synthesis.

The phoneme template data includes a sequence of digital values obtained by sampling a speech parameter P and a pitch variation Pitch expressed as a function of time t at a constant time Δt interval, and a section length T (sec.) Between the speech parameter P and the pitch pitch. .)) And can be represented by the following formula (A). In the following formula (A), t = 0, Δt, 2Δt, 3Δt,.
[Equation 1]
Template = [P (t), Pitch (t), T] (A)

  Next, the singing score data storage area 23 is singing score data representing a melody composed of a sequence of phonemes, and features of each phoneme (pronunciation timing of each phoneme, temporal change in pitch, Singing score data including attribute data (phoneme data, pronunciation timing data, pitch data, etc.) representing phonemes etc. is stored.

  FIG. 2 is a conceptual diagram showing an example of the content of singing score data. This singing score data is composed of a plurality of tracks including a phonological track and a pitch track. In the phoneme track, phoneme data representing phonemes and sounding timing data indicating the sounding start timing and sounding end timing of each phoneme are recorded. Specifically, for example, in the example shown in FIG. 2, the phoneme of “sa” phoneme is pronounced from time t1 to time t2, and the phoneme of “I” phoneme is pronounced from time t2 to time t3. It is shown that it will be. In the following, for convenience of explanation, when it is not necessary to distinguish between “sound generation start timing” and “sound generation end timing”, these will be referred to as “sound generation timing”. In the pitch track, pitch data indicating temporal changes in the fundamental frequency (pitch) of the sound to be sounded at each time is recorded.

This singing score data may be stored in advance in the singing score data storage area 23 of the storage unit 20, or generated by the control unit 10 executing a predetermined application program in response to a user operation. You may make it do. The singing score data is an example of lyric data indicating lyrics and musical score data associated with the lyrics.
FIG. 3 is a diagram illustrating an example of a screen displayed on the display unit 40 when the control unit 10 performs a singing score data generation process. The control unit 10 displays a screen illustrated in FIG. 3 and prompts the user to input singing score data. In the figure, the singing score data editing screen 600 includes an event display area 601 for displaying note data in a piano roll format. A scroll bar 606 for scrolling up and down the display screen of the event display area 601 is provided on the right side of the event display area 601. A scroll bar 607 for scrolling the display screen of the event display area 601 left and right is provided below the event display area 601.

  On the left side of the event display area 601, a keyboard display 602 (coordinate axis indicating the pitch) simulating a piano keyboard is displayed. Above the event display area 601, a bar display 604 indicating the bar position from the beginning of the music is displayed. The Reference numeral 603 denotes a piano roll display area which displays note data in a horizontally long rectangle (bar) at a time position indicated by a measure display 604 of a pitch indicated by a keyboard display 602. The left end position of the bar indicates the utterance start timing, the bar length indicates the utterance duration time, and the left end position of the bar indicates the utterance end timing.

  The user moves the mouse pointer to a position on the display screen corresponding to the desired pitch and time position and clicks to specify the utterance start position. Then, a note data bar (hereinafter referred to as “note bar”) from the utterance start position to the utterance end position is formed in the event display area 601 by a drag operation, and then the mouse is dropped. For example, in order to form the note bar 611, the mouse pointer is positioned at the first position of the first beat of the 53rd bar, the mouse is clicked, and the drag is performed after one beat.

  As described above, the user inputs the singing score data using the operation unit 30 while confirming the screen displayed on the display unit 40. The control unit 10 generates singing score data according to the signal output from the operation unit 30, and stores the generated singing score data in the singing score data storage area 23.

  Next, in the singing voice data storage area 24 of the storage unit 20, for example, voice data representing voice waveforms in the WAVE format, MP3 (MPEG Audio Layer-3) format, etc., and voice representing the singing voice sung by the user. Data (hereinafter referred to as “singing voice data”) is stored. The analysis result data storage area 25 stores analysis result data (attribute data) indicating analysis results obtained by analyzing the singing voice data with respect to a plurality of attributes. In this embodiment, the control unit 10 analyzes the singing voice data to detect the pitch, power, and spectrum of the voice, and stores data indicating the detection result in the analysis result data storage area 25 as analysis result data.

  Next, an example of a functional configuration of the speech synthesizer 100 will be described with reference to the block diagram shown in FIG. In FIG. 4, the voice synthesis unit 11, the analysis unit 12, the singing score data correction unit 13, and the confirmation sound generation unit 14 read out the computer program stored in the ROM or the storage unit 20 by the CPU of the control unit 10 and read it into the RAM. Realized by loading and executing. The CPU of the control unit 10 is an example of a voice synthesis unit 11, an analysis unit 12, a singing score data correction unit 13, and a confirmation sound generation unit 14. The speech synthesizer 11 reads the singing score data from the singing score data storage area 23, and generates speech waveform data representing the speech waveform corresponding to the singing score data from the read singing score data. More specifically, in this embodiment, the speech synthesizer 11 refers to the pitch data, pronunciation timing data, phonological data, etc. included in the singing score data, and converts the speech parameters corresponding to the pitch and phonology to the phonological template. The database 22 is read from the Timbre database 21 with reference to the database 22, and digital voice waveform data is generated using the read voice parameters. The voice synthesizer 11 performs various control processes such as singing synthesis start / stop, tempo designation, and the like. These processes are the same as those in the conventional singing synthesis technique, and detailed description thereof is omitted here. To do. Hereinafter, for convenience of explanation, the speech waveform data generated from the singing score data will be referred to as “synthesized speech data”.

  The synthesized speech represented by the synthesized speech data generated by the speech synthesizer 11 may be mechanical and unnatural. Even if it is not unnatural, there is a case where it is desired to correct the singing method (intonation etc.) desired by the user. So, in this embodiment, the control part 10 inputs the singing voice by a user, and performs the process which corrects synthetic | combination audio | voice data using this singing voice.

  The analysis unit 12 analyzes the singing voice data for a plurality of attributes including the pitch, and outputs analysis result data indicating the analysis result. In this embodiment, the analysis unit 12 analyzes audio data and detects the pitch, power, and spectrum of the audio data. For example, FFT (Fast Fourier Transform) is used for spectrum detection. The analysis unit 12 stores data indicating the analysis result in the analysis result data storage area 25.

  The singing score data correction unit 13 corrects pitch data and pronunciation timing data included in the singing score data based on the analysis result data. The singing score data correction unit 13 includes a receiving unit 131 that receives the analysis result data, a second receiving unit 132 that receives the singing score data (lyric data and score data), and the analysis result data received by the receiving unit 131 and the second The receiving unit 132 performs association with the singing score data received, and the association unit 133 associates the lyrics data with the pitch data based on the association result. More specifically, the associating unit 133 first obtains a correspondence relationship between the synthesized speech and the user singing speech based on the singing score data and the analysis result data. The voice represented by the singing voice data (hereinafter “singing voice”) and the voice represented by the synthesized voice data (hereinafter “synthesized voice”) may be shifted in time. For example, when the user sings while intentionally shifting the start or end of singing, the singing voice and the synthesized voice are shifted back and forth in time. In this way, even when the singing voice and the synthesized voice are shifted in time, the time normalization (DTW: Dynamic) is performed to expand and contract the time axis of the synthesized voice data so that they can be associated with each other. Time Warping) and align the time axes of both. As a technique for performing this DTW, DP (Dynamic programming) may be used in this embodiment.

  The associating unit 133 corrects the singing score data based on the detected difference. More specifically, the associating unit 133 corrects the pitch data and the pronunciation timing data constituting the singing score data in a direction that eliminates the difference between the synthesized voice data and the singing voice data. For the pitch, the associating unit 133 determines the pitch data value included in the singing score data based on the pitch of the singing voice data, the pitch of the synthesized voice data, and the corresponding location of the singing voice and the synthesized voice. Correction is made so that the difference between the pitch and the pitch of the synthesized speech corresponding to the pitch becomes small. Note that the amount of correction in this processing may be, for example, correcting the value of the pitch data so that the pitch of the synthesized voice matches the pitch of the singing voice, or, for example, the difference in which the difference between the two is detected You may make it correct so that it may become substantially half. Moreover, you may make it correct so that the difference of the pitch of a song voice and the pitch of a synthetic | combination voice may become below a predetermined threshold value. In short, the associating unit 133 may correct the value of the pitch data included in the singing score data so that the difference between the pitch of the synthesized speech and the pitch of the singing speech becomes small.

  In addition, the associating unit 133 corrects the value of the pronunciation timing data included in the singing score data so that the difference between the pronunciation timing detected from the singing voice data and the pronunciation timing detected from the synthesized voice data becomes small. . This correction amount is also the same as the pitch correction described above, and the value of the sound generation timing data may be corrected so that the sound generation timing of the synthesized speech matches the sound generation timing of the singing sound. The associating unit 133 updates the stored content of the singing score data storage area 23 with the singing score data obtained by correcting each attribute data. The singing score data stored in the singing score data storage area 23 is referred to when the speech synthesizer 11 performs speech synthesis processing.

  By the way, after the user inputs the singing voice and before the synthesized voice corrected by the inputted singing voice is reproduced, the processing by the singing score data correcting unit 13 and the voice synthesizing unit 11 described above is required. . At this time, since the processing performed by the singing score data correction unit 13 and the speech synthesis unit 11 requires a certain amount of processing time, the user needs to wait until the processing is completed. When the user repeatedly corrects the synthesized speech, it is necessary to wait each time the correction is made in order to check the corrected speech, and the synthesized speech editing process may not be performed smoothly. . Therefore, in the present embodiment, a sound signal based on the pitch indicated by the analysis result data (hereinafter referred to as “confirmation sound signal”) is generated and output by the confirmation sound generation unit 14 according to the operation by the user, and the analysis result of the input speech It is easy to confirm.

  The confirmation sound generation unit 14 receives the analysis result data generated by the analysis unit 12, and generates a confirmation sound signal having periodicity based on the pitch indicated by the received analysis result data. In this embodiment, the confirmation sound generator 14 generates a sine wave having a frequency corresponding to the pitch indicated by the analysis result data. The confirmation sound generation unit 14 supplies the generated confirmation sound signal to the sound processing unit 60 and emits a sound corresponding to the generated confirmation sound signal (hereinafter referred to as “confirmation sound”) from the speaker 62.

<Operation>
FIG. 5 is a flowchart showing the flow of the synthesized speech correction process performed by the speech synthesizer 100. When the user gives an instruction to edit the synthesized voice via the operation unit 30 (step S100; Yes), the control unit 10 first waits for the singing voice to be input (step S102; No). When the singing voice is input by the user (step S102; Yes), the input singing voice is analyzed, and analysis result data indicating the analysis result is generated (step S104).

  Next, the control unit 10 determines whether or not to reproduce the confirmation sound according to the user operation (step S106). In this process, for example, the control unit 10 may display a button for generating a confirmation sound on the display unit 40, and may determine that the confirmation sound is reproduced when the button is clicked. When it is determined not to reproduce the confirmation sound (step S106; NO), the control unit 10 proceeds to the process of step S110 without performing the process of step S108. On the other hand, when it is determined that the confirmation sound is reproduced (step S106; YES), the control unit 10 generates a confirmation sound signal having periodicity based on the pitch indicated by the analysis result data (step S108). The sound represented by the confirmation sound signal is emitted from the speaker 62.

  Since the analysis result of the sound has a subtle change in pitch, it may be difficult to grasp what kind of sound the subtle change is actually without listening to the actual sound. Although the confirmation sound reproduced in step S108 is not a synthesized voice to be finally generated, it is a sound in which the pitch of the synthesized voice to be generated is represented. Therefore, the user can listen to the reproduced sound. It is possible to intuitively understand what kind of sound is generated. At this time, the process of generating the confirmation sound (that is, the process performed by the confirmation sound generation unit 14) is compared with the process of generating the synthesized speech (that is, the process performed by the singing score data correction unit 13 and the speech synthesis unit 11 described above). Therefore, since the amount of calculation is small and processing is performed in a short time, the user does not have to wait for each time to check the result of speech analysis.

  Returning to the description of FIG. The user can select whether to generate synthesized speech or re-input singing speech. The user selects whether to generate synthesized speech by operating the operation unit 30, and the control unit 10 determines whether to generate synthesized speech in accordance with the user's operation (step S110). When it is determined that the synthesized speech is to be generated (step S110; Yes), the control unit 10 performs the processes of the singing score data correction unit 13 and the speech synthesis unit 11 described above to generate the synthesized speech data (step S112). ). That is, the control unit 10 corrects the singing score data based on the analysis result data, and generates synthesized speech data from the corrected singing score data with reference to the Timbre database 21 and the phonological template database 22. On the other hand, when it is determined that the singing voice is to be input again (step S112; No), the control unit 10 returns to the process of step S100 and waits for the input of the correction instruction.

<Modification>
The above embodiment can be modified as follows. In addition, you may implement the following modifications suitably combining.

<Modification 1>
In the above-described embodiment, the control unit 10 generates a sine wave having a frequency corresponding to the pitch indicated by the analysis result data as the confirmation sound. However, the confirmation sound signal generated by the control unit 10 is not limited to this, for example, A sine wave having an amplitude corresponding to the frequency corresponding to the pitch indicated by the analysis result data and the volume (power) indicated by the analysis result data may be generated. For example, the control unit 10 may distort the waveform by performing a predetermined modulation process on a sine wave having a frequency corresponding to the pitch indicated by the analysis result data. Further, for example, the control unit 10 uses a sound signal obtained by synthesizing a frequency component corresponding to the pitch indicated by the analysis result data and a specific harmonic component such as a second harmonic or a third harmonic of the frequency component as the confirmation sound signal. May be. Further, for example, the control unit 10 may generate a confirmation sound signal by synthesizing the harmonic component F0 up to the nth harmonic of the frequency component corresponding to the pitch indicated by the analysis result using the following equation (B). . In the following formula (B), POW is power, a is a constant or a value imitating a formant from peak information of a spectrum which is an analysis result of singing voice data. When a is a constant, a confirmation sound signal like a nose song is generated, and when a value imitating a formant is used as a, a confirmation sound signal similar to the user's singing sound is generated.
[Equation 2]
Σsin (n · F0) * (a · POW) (B)

  As described above, the confirmation sound signal may be a sine wave having a frequency corresponding to the pitch indicated by the analysis result data. For example, a frequency component corresponding to the pitch indicated by the analysis result data and its harmonic component are synthesized. In short, the control unit 10 may generate a sound signal having periodicity based on the pitch indicated by the analysis result data. In the above-described embodiment, a sine wave is used as the confirmation sound signal. However, the confirmation sound signal is not limited to this and may be a sound signal having a simple waveform such as a triangular wave or a rectangular wave. Further, a confirmation sound signal representing the tone color of the musical instrument may be generated using a known technique such as musical instrument sound synthesis. The confirmation sound signal may be any sound signal that is generated based on the analysis result data indicating the analysis result of the singing voice data and is generated by a process with a light processing load.

<Modification 2>
In the above-described embodiment, the control unit 10 corrects the pitch data and the pronunciation timing data included in the singing score data, but the attribute data to be corrected is not limited to this. For example, the control unit 10 may detect a sound quality / voice quality difference and correct the sound quality / voice quality. In this case, the singing score data includes sound quality data and voice quality data indicating the sound quality and voice quality, and the control unit 10 detects formants from the singing voice data and the synthesized voice data, and the difference between the detected formants is small. As such, sound quality data and voice quality data may be corrected.

  As described above, the attribute data representing the sound attribute to be corrected by the control unit 10 may be pitch data or pronunciation timing data indicating the temporal change of the pitch shown in the above-described embodiment, or phoneme data. Or sound quality data or voice quality data. As another example, for example, data representing the velocity (strongness) of sound, or data representing the form of vibrato may be used. As described above, the attribute data corrected by the control unit 10 may be any data as long as it represents a voice attribute.

  Moreover, although the control part 10 demonstrated the aspect which corrects singing score data based on the analysis result of song voice data in the above-mentioned embodiment, it is not restricted to this, The analysis result itself of song voice data is used as song score data. Speech synthesis may be performed.

  In the above-described embodiment, the control unit 10 analyzes the singing voice data with respect to a plurality of attributes including the pitch, and generates analysis result data indicating the analysis result. However, the control unit 10 analyzes a plurality of attributes. Not limited to this, only the pitch may be analyzed, and attribute data indicating the analyzed pitch may be generated.

<Modification 3>
In the above-described embodiment, the control unit 10 reads the singing score data from the singing score data storage area 23. However, the mode in which the speech synthesis unit 11 acquires the singing score data is not limited to this, for example, the Internet or the like. The singing score data may be received via the communication network, and for example, the user performs an operation for inputting the singing score data using the operation unit 30, and the control unit 10 performs the operation from the operation unit 30. Singing score data may be generated according to the output signal, and any data may be used as long as the control unit 10 acquires the singing score data.

  In the above-described embodiment, the singing score data is used as the lyric data and the score data. However, the structure of the lyric data and the score data is not limited to that exemplified in the above-described embodiment. Data having any structure may be used as long as the correspondence between the notes and the lyrics and the attributes of the notes can be specified. In the embodiment, the example in which the lyrics (character string) and the score data are separate data sets has been described. However, the lyrics may be a part of the score data.

  The details of the speech synthesis process are not limited to those described in the embodiment. As long as a note and a phonetic symbol (character) are given, any processing may be used as long as it synthesizes a sound corresponding to the note and the phonetic symbol.

Moreover, in the above-mentioned embodiment, it was set as the structure which corrects a synthetic | combination voice by re-inputting a singing voice (refer FIG.5 S112-step S102), However, The aspect of correction | amendment of a synthetic | combination voice is limited to this. Instead, for example, the user may perform an operation for correcting the singing score data using the operation unit 30, and the control unit 10 may correct the singing score data according to the operation content of the operation unit 30. .
Moreover, in the above-mentioned embodiment, although the control part 10 analyzed the song voice of a singer, it may replace with a song voice of a singer, and may evaluate the performance sound of the musical instrument by a player. The “speech” referred to in the present embodiment includes various sounds such as a sound generated by a person and a performance sound of a musical instrument.

<Modification 4>
In the above-described embodiment, the control unit 10 analyzes the singing voice data and generates the analysis result data indicating the analysis result. However, the control unit 10 is not limited to generating the analysis result data, and other devices (for example, The analysis result may be acquired from a server device connected via a communication network. In this case, the analysis result data acquired by the control unit 10 may be data indicating a plurality of attributes including the pitch, or may be data indicating only the pitch.

<Modification 5>
The hardware configuration of the speech synthesizer 100 is not limited to that described with reference to FIG. As long as the function shown in FIG. 4 can be implemented, the speech synthesizer 100 may have any hardware configuration. For example, the speech synthesizer 100 may have dedicated hardware (circuit) corresponding to each of the functional elements shown in FIG.

<Modification 6>
In the above-described embodiment, two or more devices connected by a communication network share functions related to the speech synthesizer 100 of the above-described embodiment, and a system including these devices is a speech synthesizer of the same embodiment. 100 may be realized. For example, a computer device including a microphone, a speaker, a display device, an operation unit, and the like and a server device that performs voice analysis processing may be configured as a system connected via a communication network. In this case, for example, the computer apparatus converts the sound collected by the microphone into an audio signal and transmits it to the server apparatus, and the server apparatus analyzes the received audio signal and transmits the analysis result to the computer apparatus. May be.

<Modification 7>
In addition to the speech synthesizer, the present invention can be understood as a method for realizing these and a program for causing a computer to realize a speech synthesis function. Such a program may be provided in the form of a recording medium such as an optical disk storing the program, or may be provided in the form of being downloaded to a computer via the Internet or the like and installed and used.

DESCRIPTION OF SYMBOLS 10 ... Control part, 20 ... Storage part, 21 ... Timbre database, 22 ... Phoneme template database, 23 ... Singing score data storage area, 24 ... Singing voice data storage area, 25 ... Analysis result data storage area, 30 ... Operation part, DESCRIPTION OF SYMBOLS 40 ... Display part, 60 ... Speech processing part, 61 ... Microphone, 62 ... Speaker, 70 ... Bus, 100 ... Speech synthesizer

Claims (4)

A receiving unit for receiving the attribute data indicating an attribute containing the pitch and volume of the audio data,
A speech synthesis unit that generates synthesized speech based on the previous SL attribute data receiving unit has received,
A sound signal generator for generating a sound signal having periodicity based on a pitch and a volume indicated by the attribute data received by the receiver before the voice synthesizer generates a synthesized voice; And a sound signal generation unit that generates a sound signal having periodicity by processing in a shorter time than processing for generating synthesized speech . A second receiving unit for receiving lyric data indicating lyrics and score data associated with the lyrics;
Correspondence between the pitch indicated by the attribute data received by the receiving unit and the score data received by the second receiving unit, and correspondence between the lyrics data and pitch data representing the pitch based on the association result And an associating unit for attaching,
The speech synthesis unit generates synthesized speech based on the attribute data received by the reception unit and the lyric data and pitch data associated by the association unit. Speech synthesizer. The speech synthesizer according to claim 1 or 2, further comprising: a speech analysis unit that analyzes the speech data for attributes including pitch and volume and supplies attribute data indicating an analysis result to the reception unit. . On the computer,
A receiving function of receiving the attribute data indicating an attribute containing the pitch and volume of the audio data,
A speech synthesis function for generating a synthesized speech based on the previous SL received attribute data,
A sound signal generating function for generating a sound signal having periodicity based on a pitch and a volume indicated by the received attribute data before the voice synthesizing function generates a synthesized voice, wherein the voice synthesizing function is a synthesized voice; A program for realizing a sound signal generation function for generating a sound signal having periodicity by processing in a shorter time than processing for generating the sound .

Images