JP2010085737A - Encoding apparatus and information communication system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To superimpose watermark information to sound in a balanced state between sureness in the detection of the watermark information and the deterioration in sound quality. <P>SOLUTION: An encoding apparatus 100 allows a speaker 500 to discharge an adjustment sound and calculates an adjustment value for determining a superimposition state of watermark information on the basis of a sound collection result of a microphone 600 collecting the adjustment sound. An analysis part 140 calculates the adjustment value and a superimposition part 130 superimposes the watermark information to sound information by using the adjustment value. The superimposition part 130 superimposes the watermark information to the sound information after adjusting superimposition state so as to easily decode the sound information. Concretely, an adjustment part 440 adjusts a frequency band or the like for superimposing a sound pressure level of a sound expressed by the sound information or the watermark. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a technique for superimposing watermark information on sound.

As a technique for ensuring the detection of watermark information added by the digital watermark technique, there is one described in Patent Document 1, for example. Patent Document 1 describes a technique for superimposing the same watermark information (additional information) on the original information (main information signal) by a plurality of methods.
JP 2002-101397 A

In the technique described in Patent Document 1, watermark information is redundantly added. In this case, the more redundancy is added to the superposition, the higher the possibility that the original information is degraded.
On the other hand, an object of the present invention is to superimpose watermark information on sound while balancing the certainty of detection of watermark information and deterioration of sound quality.

  The encoding apparatus according to the present invention includes a superimposing unit that superimposes watermark information on sound information according to an aspect, and a sound emitted by the sound emitting unit obtained by collecting the sound emitted by the sound emitting unit. Acquisition means for acquiring characteristic information indicating the characteristics of the image, determination means for determining a superposition mode by the superimposition means in accordance with the characteristic information acquired by the acquisition means, and the superposition means by the mode determined by the determination means Comprises output means for outputting sound information on which watermark information is superimposed so as to be supplied to the sound emitting means.

  In the encoding device according to the present invention, the acquisition unit decodes information representing the sound collected by the sound collection unit that collects the sound emitted by the sound emission unit or the watermark information decoded from the sound. You may employ | adopt the structure which acquires a result as the said characteristic information.

  An information communication system according to the present invention includes: a transmitter including the encoding device and sound emitting means; the sound collecting means; and decoding means for decoding the watermark information from the sound collected by the sound collecting means. And a receiver.

  According to the present invention, it is possible to superimpose watermark information on sound while balancing the certainty of detection of watermark information and deterioration of sound quality.

[First Embodiment]
FIG. 1 is a diagram showing a configuration of an encoding apparatus according to the first embodiment of the present invention. The encoding apparatus 100 acquires sound information and watermark information by an acquisition unit (not shown), superimposes the watermark information on the sound information, and outputs the information to the speaker 500. Further, the encoding apparatus 100 further includes acquisition means for acquiring sound information (hereinafter referred to as “sound collection information”) representing sound collected from the microphone 600 that collects the adjustment sound emitted by the speaker 500. And the superposition mode is changed based on the sound collection information. The encoding device 100 includes a command supply unit 110, an adjustment sound generation unit 120, a superimposition unit 130, an analysis unit 140, and an output unit 150 in addition to the acquisition unit that acquires the various types of information.

  The command supply unit 110 generates a command that triggers an adjustment process, which will be described later, and supplies the command to the adjustment sound generation unit 120 and the analysis unit 140. The command supply unit 110 may generate a command in response to an operation by a remote operation device (so-called remote controller), or may generate a command at a predetermined timing (when the power is turned on or at a predetermined time interval).

  The adjustment sound generator 120 supplies sound information representing the sound (adjustment sound) emitted from the speaker 500 for adjustment to the output unit 150. In the following, the sound information representing the adjusted sound is referred to as “adjusted sound information” and is distinguished from other sound information. The adjustment sound is a sound that is emitted in order to obtain sound transmission characteristics by the speaker 500, and is, for example, a sound whose frequency changes continuously with time (so-called sweep sound).

The superimposing unit 130 acquires sound information and watermark information, and performs a process of superimposing the watermark information on the sound information. The method for superimposing the watermark information is not particularly limited. For example, among the modulation methods using sound waves as a carrier wave, an amplitude modulation method or a phase modulation method can be applied. The amplitude modulation method may be one that modulates the amplitude of a specific frequency band according to watermark information, or divides the entire emitted frequency band into several bands, and each of the divided bands. Modulation corresponding to watermark information may be performed. The superposition method of this embodiment is assumed to be the latter amplitude modulation method.
In addition, the superimposing unit 130 acquires an adjustment value from the analysis unit 140, and determines a superposition mode according to the adjustment value. That is, the superimposing unit 130 is an example of a determining unit in the present invention. It is assumed that there is a predetermined standard mode for the superposition mode.

  The analysis unit 140 acquires sound collection information triggered by the acquisition of the command supplied by the command supply unit 110, and calculates an adjustment value for adjusting the superposition mode. The adjustment value is calculated based on the sound emitted from the speaker 500, and is information reflecting the characteristics of the sound emitted from the speaker 500. That is, the adjustment value is an example of characteristic information of the present invention.

  The output unit 150 outputs the supplied sound information or adjustment sound information to the speaker 500. When the adjustment process is performed, the output unit 150 outputs the adjustment sound information, and then outputs the sound information that is superimposed in a manner adjusted based on the adjustment process performed most recently. Note that the output unit 150 may be configured to output so that sound information is finally supplied to the speaker 500. Therefore, another component (such as an amplifier) may be interposed between the output unit 150 and the speaker 500.

The configuration of the encoding device 100 of the present embodiment is as described above. In the present embodiment, the adjustment process is performed so that the sound is emitted in such a manner that the watermark information is accurately decoded before the sound on which the watermark information is superimposed is emitted. The adjustment process of the present embodiment is a process that is executed when the command supply unit 110 generates a command.
In the adjustment process, the encoding apparatus 100 first causes the speaker 500 to emit the adjustment sound. The microphone 600 collects the adjustment sound, and supplies the collected sound information to the analysis unit 140 of the encoding device 100.

  Based on the acquired sound collection information, the analysis unit 140 analyzes characteristics related to sound emission and sound collection of the adjustment sound, and calculates an adjustment value based on the analysis result. The analysis unit 140 analyzes the characteristics related to sound emission and sound collection of the adjusted sound by specifying how much the sound collection information differs from a predetermined reference. Here, the reference is determined based on the adjustment sound information, for example. For example, when the adjustment sound information is used as a reference, the analysis unit 140 has a sound pressure level in a certain frequency band of the adjustment sound represented by the sound collection information that is relatively smaller than other bands or lower than the reference level. Is also smaller than a predetermined value, an adjustment value indicating that the sound pressure level in the frequency band is small is calculated. Alternatively, the analysis unit 140 may calculate an adjustment value that indicates what adjustment is necessary in the superimposition unit 130 based on the analyzed characteristics.

  The analysis unit 140 calculates an adjustment value indicating that when the noise included in the adjustment sound represented by the collected sound information is small or the sound pressure level of the adjustment sound is sufficiently high. May be. In this case, the reference noise level and the reference adjustment sound pressure level are determined in advance.

  When the superimposing unit 130 acquires the adjustment value, the superimposing unit 130 determines the superimposing mode according to the adjustment value. For example, when the superimposing unit 130 obtains an adjustment value indicating that the sound pressure level in a certain frequency band is small, the superimposing unit 130 performs superimposition after increasing the sound pressure level corresponding to the frequency band of the sound information from the standard. Or the other frequency band is used without using the frequency band of the sound information for superimposing the watermark information. In the latter case, when adaptively controlling the band used for information superposition and the band not used according to the adjustment value, information that can identify that the information is superposed (for example, for each band) It is desirable to include (periodic synchronization sound) in the sound information.

  In addition, when the superimposition unit 130 acquires an adjustment value indicating that the noise included in the adjustment sound represented by the sound collection information is low, or an adjustment value indicating that the sound pressure level of the adjustment sound is sufficiently large, You may superimpose, after making a pressure level smaller than a standard. Alternatively, the superimposing unit 130 changes the superposition mode from the standard mode to the other mode when such adjustment values are acquired when there is another mode with less deterioration in sound quality than the standard superposition mode. It is also possible to perform superposition by switching to this mode. In this case, the other mode may be one in which the amount of information per unit time of the watermark information is larger than the standard mode.

  The sound information on which the watermark information is superimposed in the manner adjusted as described above is converted into sound by the speaker 200 and emitted. This sound is picked up by a decoding device (not shown), and watermark information is decoded (demodulated). The decoding device decodes the watermark information by a method according to a predetermined superposition method. Note that the sound collection means included in the decoding device may be the microphone 600 described above.

  As described above, the encoding apparatus 100 according to the present embodiment determines a superposition mode based on the sound collection result of the adjustment sound. Therefore, the encoding apparatus 100 eliminates the factors that make it difficult to detect and decode the watermark information due to the situation, such as the characteristics of the speaker 500 itself and ambient noise, for example. Thus, it is possible to change the superposition mode. Also, the encoding apparatus 100 can change the superposition mode even when sufficiently accurate detection of watermark information can be expected, such as when there is little noise.

  In general, in sound information, the sound quality tends to deteriorate as the superposition mode in which watermark information is more easily detected. That is, in the audio watermark, it can be said that there is a trade-off relationship between the certainty of detection of watermark information and the deterioration of sound quality. According to the encoding apparatus 100 of the present embodiment, not only the superimposition mode is changed in a direction that more reliably detects watermark information, but also the superposition mode can be changed in a direction that improves sound quality. Thus, it is possible to perform superimposition in an optimal manner in which the certainty of watermark information detection and sound quality degradation are balanced. Therefore, according to the encoding apparatus 100 of the present embodiment, if there is a lack of sound quality, the lack of sound quality is compensated, and if there is a lack of detection accuracy, the detection is changed to a mode that is easier to detect. It is possible to adaptively change the superposition mode in accordance with.

[Second Embodiment]
FIG. 2 is a diagram showing a configuration of an encoding apparatus according to the second embodiment of the present invention. In the present embodiment, the command supply unit 210, the adjustment sound generation unit 220, the analysis unit 250, the output unit 260, the speaker 500, and the microphone 600 shown in the figure have the same functions as the components having the same names in the first embodiment. Have. Therefore, description of these components is omitted. Descriptions of matters common to the present embodiment and the first embodiment are omitted as appropriate.

  The sound information of the present embodiment includes at least information indicating a pitch (pitch). The sound information may include information indicating the intensity and length of the sound. As sound information of the present embodiment, for example, MIDI (Musical Instrument Digital Interface: registered trademark) data can be used. When MIDI data is used as sound information, the information indicating the pitch is a note number. Further, the sound information may be one that continuously emits a sound with a certain pitch, or may be one whose pitch changes with time like a musical piece. When the sound information is like the former, the sound information may not be acquired from the outside of the encoding apparatus 200 but may be acquired by reading out data stored therein.

  In addition, the superimposing method of the present embodiment represents watermark information by shifting the frequency and phase of a sine wave signal from a reference value in the case where a sound of a certain tone color is expressed by synthesizing a plurality of sine wave signals. Is embedded. Here, the “deviation” from the reference value is determined according to the watermark information (details will be described later).

  The encoding device 200 includes a synthesized sound generation unit 230 and an adjustment unit 240. The encoding device 200 includes storage means such as a memory and stores a timbre database DB1. The timbre database DB1 stores timbre data. The timbre data is a set of timbre parameters for adding a timbre to the sound represented by the sound information. The tone color parameters here include the frequency, amplitude, phase, and the like of overtones. The timbre data may be given as data representing the timbre of a predetermined instrument such as “snare drum” or “piano”, for example. In this case, the timbre data is stored for a plurality of timbres in the timbre database DB1, and which timbre to use is selected by a button or the like provided on the remote control device or the encoding device 200. Good.

  The synthesized sound generation unit 230 superimposes watermark information on the sound information and supplies it to the output unit 260. The watermark information superimposed by the synthesized sound generation unit 230 is information after adjustment by the adjustment unit 240. The synthesized sound generation unit 230 and the adjustment unit 240 constitute a determination unit according to the present invention in cooperation.

FIG. 3 is a diagram illustrating a main configuration of the synthesized sound generation unit 230. As shown in the figure, the synthesized sound generating unit 230 includes an oscillator 231 _1, 231 _2, 231 _3, ..., and 231 _n, an amplifier 232 _1, 232 _2, 232 _3, ..., and 232 _n, an adder 233 Is provided. Here, the subscripts (1 to n) of the reference numerals are attached to distinguish each. Therefore, when it is not necessary to distinguish each of the oscillators 231 _{1 to} 231 _n or the amplifiers 232 _{1 to} 232 _n , their subscripts are omitted and expressed as “oscillator 231” or “amplifier 232”. The number of oscillators 231 and amplifiers 232, that is, the value of n is an arbitrary natural number.

The oscillator 231 oscillates at a specified frequency and generates a sine wave signal representing a sine wave. The amplifier 232 is an amplifier whose gain is variable, and amplifies the level of the sine wave signal oscillated by the corresponding oscillator 231. The adder 233 adds the sine wave signals amplified by the amplifier 232 and outputs the result as an acoustic signal representing one waveform.
The frequencies assigned to the oscillators 231 ₁ , 231 ₂ , 231 ₃ ,..., 231 _n are different from each other. The gains of the amplifiers 232 ₁ , 232 ₂ , 232 ₃ ,... 232 _n are determined for each designated frequency. The frequency, phase, and gain are determined based on the timbre parameters adjusted by the adjustment unit 240.

  The adjustment unit 240 acquires watermark information and timbre parameters, and changes the timbre parameters according to the watermark information. This change corresponds to the above-described “deviation”. In addition, the adjustment unit 240 changes the timbre parameter according to the acquired adjustment value at the time of this change. Since this change corresponds to the adjustment in the first embodiment, it will be distinguished from the former change by “adjustment” below.

  The configuration of the encoding apparatus 200 of the present embodiment is as described above. Also in the present embodiment, as in the first embodiment, an adjustment process for calculating an adjustment value by emitting an adjustment sound is performed in advance, and watermark information is superimposed in a manner corresponding to the adjustment value.

FIG. 4 is a view for explaining the principle of superposition in the present embodiment. Now, as sound information, data of a sound having a pitch, that is, a frequency (fundamental frequency) of f ₀ is given, and data of frequencies f ₁ , f _2, and f ₃ and amplitude levels corresponding thereto are provided as tone color parameters. Suppose you are given. The amplitude levels of the frequencies f ₁ , f ₂ and f ₃ are relatively determined based on the sound level of the frequency f ₀ . When such information is given, if the synthesized sound generator 230 oscillates, amplifies and synthesizes with these frequencies, the waveforms of the frequencies f ₀ , f ₁ , f ₂ and f ₃ are synthesized. A sound signal is obtained. The frequency spectrum of this acoustic signal is as shown in FIG.
Here, it is desirable that the frequencies f ₁ , f _2, and f ₃ have a harmonic relationship with respect to the fundamental frequency f ₀ , but they do not necessarily have to be harmonics.

When such sound information and timbre parameters are given, it is assumed that watermark information with a 3-digit bit value of “101” is given. When such watermark information is supplied, the adjustment unit 240 sets the first digit bit value “1” to the frequency f ₁ and the second digit bit value “0” to the frequency f ₂ and the third digit. Are assigned to the frequency f ₃ . The bit values “0” and “1” are each given a shift amount of a predetermined frequency, and the adjustment unit 240 changes the timbre parameter according to the shift amount. That is, at this time, the adjustment unit 240 shifts the frequency of the timbre parameter by the shift amount corresponding to the watermark information. Here, when the shift amount corresponding to the bit value of “0” is “Δ ₀ ” and the shift amount corresponding to the bit value of “1” is “Δ ₁ ”, the frequency spectrum of the acoustic signal after the change is As shown in FIG.

Note that it is desirable that Δ ₀ and Δ _{1 be} sufficiently smaller than the above-described difference in frequency (that is, | f ₁ −f ₂ | or | f ₂ −f ₃ |). This is because if Δ ₀ and Δ ₁ are sufficiently small, it is possible to prevent the timbre from being perceived by hearing. However, if Δ ₀ and Δ ₁ are small, it becomes difficult to decode the watermark information. Therefore, the actual shift amount is appropriately determined in consideration of the change in tone color and the decoding success rate.

It can be said that the frequency spectrum in FIG. 4B is a frequency spectrum when adjustment according to the adjustment value is not performed. That is, the adjustment unit 240 makes further changes to the timbre parameters when performing adjustment according to the adjustment value. Specifically, the adjustment unit 240 changes at least one of the amplitude levels of the frequencies f ₀ , f ₁ , f _2, and f ₃ and the frequency shift amount. The frequency spectrum of the acoustic signal when such adjustment is performed is, for example, as shown in FIG. Here, “α” added to the shift amount is a value determined according to the adjustment value. The adjustment for increasing the amplitude level may be performed only in a specific frequency band, or may be performed over the entire frequency band to be emitted.

  Here, assuming that the frequency spectrum of FIG. 4B is a standard, it can be said that the frequency spectrum of FIG. 4C has a frequency shift amount changed by “α” with respect to the standard. The value of “α” can be varied depending on the adjustment value. For example, the value of “α” may be increased to ensure the detection of watermark information, and the value of “α” may be decreased to suppress deterioration in sound quality. If sufficiently accurate detection of watermark information can be expected, the value of “α” may be a negative value, and the frequency shift amount may be smaller than the standard.

In this example, the shift amount can be adjusted when the shift amount itself has no information. In the example shown here, when the frequency is shifted in the positive direction, it is “1”, and when the frequency is shifted in the negative direction, it is “0”. The magnitude (as an absolute value) has no information. Therefore, as shown in FIG. 4C, even if the shift amount is changed to “Δ ₀ + α” or “Δ ₁ + α”, the watermark information is not lost by the change. On the other hand, for example, when the shift amount corresponding to the bit value of “0” and the shift amount corresponding to the bit value of “1” are both in the positive direction and the values are distinguished from each other, the shift amount Since there is a possibility that information may be lost due to a change in the shift amount, the shift amount may not be adjusted.

  The sound information on which the watermark information is superimposed in the manner adjusted as described above is converted into sound by the speaker 200 and emitted. This sound is picked up by a decoding device (not shown), and watermark information is decoded (demodulated). The decoding device decodes the watermark information by a method according to a predetermined superposition method. The decoding apparatus stores data corresponding to the above-described timbre parameters, that is, reference data, and can obtain watermark information by specifying a deviation of the collected sound from the reference. The decoding apparatus can specify a deviation from the reference by converting the acoustic signal into a set of frequency components by FFT (Fast Fourier Transform), for example. When a plurality of timbres are selected from the candidates, the decoding device includes means for recognizing the timbre selected by the encoding device 200 by obtaining the data as data from the outside.

  As described above, in the present embodiment, the same effects as those in the first embodiment can be obtained, although the manner of superimposition and adjustment is different.

[Modification]
The present invention can be implemented in a form different from the above-described embodiment. The following modifications are examples of modifications applicable to the present invention. In addition, these modifications may be implemented combining each as needed.

(1) Modification 1
The encoding apparatus according to the present invention decodes the sound information that is the sound collection result of the sound with the watermark information superimposed thereon or the collected sound, instead of the configuration in which the adjustment sound is emitted and the sound collection result is acquired. A configuration for acquiring the decoded result may be provided. These pieces of information also indicate the characteristics of the emitted sound and correspond to an example of the characteristic information in the present invention.

  FIG. 5 is a diagram showing a configuration of the information communication system according to the present example. The information communication system 10 includes an encoding device 300, a speaker 500, a microphone 600, and a decoding device 700. The encoding device 300 and the speaker 500 cooperate to constitute a watermark information transmitter, and the microphone 600 and the decoding device 700 cooperate to constitute a watermark information receiver. The encoding device 300 has a configuration in which the command supply unit 110 and the adjustment sound generation unit 120 are removed from the encoding device 100 of the first embodiment, but the function of the analysis unit 140 is different from that of the first embodiment. The encoding apparatus 300 superimposes predetermined watermark information on sound information and causes the speaker 500 to emit sound. The decoding device 700 acquires sound information from the microphone 600 and decodes the watermark information according to a predetermined superposition method. That is, the decoding device 700 is an example of decoding means in the present invention. In addition, the decoding device 700 supplies result data that is information indicating the decoding result of the watermark information to the analysis unit 140 of the encoding device 300. The result data may be information that directly represents the decoding result, or may be information that represents a location where the decoding has succeeded and a location where the decoding has failed. Alternatively, the result data may be information indicating the success rate of decoding and the occurrence rate of errors.

  The analysis unit 140 calculates an adjustment value based on the result data. The analysis unit 140 includes means for comparing the predetermined watermark information (original information) output from the encoding device 300 with the result data that is the transmission result, and analyzes how correctly the watermark information has been transmitted. It may be a configuration. When the watermark information is not correctly transmitted, the analysis unit 140 calculates an adjustment value that makes transmission more reliable.

  FIG. 6 is a diagram showing another configuration of the information communication system according to the present example. The information communication system 10 shown in FIG. 5 is a modification of the first embodiment, but the information communication system 20 is a modification of the second embodiment. The information communication system 20 includes an encoding device 400, a speaker 500, a microphone 600, and a decoding device 700. The encoding device 400 has a configuration in which the command supply unit 210 and the adjustment sound generation unit 220 are removed from the encoding device 200 of the second embodiment, but the function of the analysis unit 250 is different from that of the second embodiment. The encoding device 400 is obtained by applying the characteristic function of the encoding device 300 of FIG. 5 to the encoding device 200 of the second embodiment.

  According to the configuration of this example, it is possible to output sound information on which watermark information is superimposed without emitting an adjustment sound. In other words, according to the configuration of this example, the output of the sound information on which the watermark information is superimposed is not hindered by the output of the adjustment sound information. Therefore, according to the configuration of this example, it is possible to make adjustments as necessary as needed even while sound information is being output.

(2) Modification 2
The adjustment in the present invention is not limited to changing the superposition mode in a predetermined superposition method, and may be performed by changing the superposition mode by selecting an optimum method from a plurality of superposition methods. Good. For example, when a certain superimposition method is used as a standard method, when the sound (noise) other than the sound emitted from the speaker increases, it is more resistant to noise than the standard method (less susceptible to noise). ) On the other hand, when the superimposition method is selected, when the noise is reduced, a superimposition method with better sound quality (small deterioration in sound quality) is selected.

(3) Modification 3
In the sound information in the present invention, when a sound with a temporal change is regular, such as an electronic siren or a bell, a parameter related to time may be used as a timbre parameter. Such parameters include, for example, attack (the time from when a sound is produced until it reaches the maximum level), decay (the time until the volume changes from the maximum level to the sustain level), and hold (the time for maintaining the volume at a certain level). ), Release (the time from when the volume decreases to 0). When modulation is performed using such parameters, a so-called envelope changes according to information.

The figure which shows the structure of the encoding apparatus of 1st Embodiment. The figure which shows the structure of the encoding apparatus of 2nd Embodiment. The figure which shows the structure of a synthetic sound production | generation part The figure for demonstrating the principle of the superimposition in 2nd Embodiment. The figure which shows the structure of the information communication system which is a modification of 1st Embodiment. The figure which shows the structure of the information communication system which is a modification of 2nd Embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100, 200, 300, 400 ... Coding apparatus, 110, 210 ... Command supply part, 120, 220 ... Adjustment sound generation part, 130, 240 ... Adjustment part, 140, 250 ... Analysis part, 150, 260 ... Output part, 230: Synthetic sound generation unit, 500 ... Speaker, 600 ... Microphone, 700 ... Decoding device

Claims (3)

Superimposing means for superimposing watermark information on sound information according to an aspect;
Obtaining means for obtaining characteristic information indicating characteristics of the sound emitted by the sound emitting means obtained by collecting the sound emitted by the sound emitting means;
Determining means for determining a superposition mode by the superimposing means according to the characteristic information obtained by the obtaining means;
An encoding device comprising: output means for outputting the sound information on which the watermark information is superimposed by the determining means so that the watermark information is supplied to the sound emitting means. The acquisition unit acquires, as the characteristic information, information representing the sound collected by the sound collection unit that collects the sound emitted by the sound emission unit or a decoding result obtained by decoding the watermark information from the sound. The encoding apparatus according to claim 1. A transmitter comprising the encoding device according to claim 1 or 2 and sound emission means;
An information communication system comprising: the sound collection unit according to claim 2; and a receiver including a decoding unit that decodes the watermark information from the sound collected by the sound collection unit.

Images