CN1236652C - Stereo sound effect production method - Google Patents

Abstract

The invention discloses a method for generating stereo sound effect (3 Daudio), which can synthesize a binaural sound effect from a monaural sound source by using a monaural head corresponding transform algorithm (HRTF) technology; comprises the following steps: measuring double ears to obtain a conversion function of the two ears; recording a set of HRTF coefficients of one ear through a monaural HRTF measurement operation to establish a monaural head correspondence transformation algorithm (HRTF) database; calculating the interaural sound time difference (ITD) compensation curve according to the conversion function of the two ears; and converting an externally input monaural signal into a stereo sound effect signal according to the two monaural HRTF databases and the ITD compensation curve.

Description

Stereo sound effect production method

technical field

The present invention relates to the processing of stereo sound effect, especially a kind of stereo sound effect generation method (implementation method of 3D audio) that can utilize monaural head corresponding conversion algorithm (HRTF) technology to synthesize a binaural sound effect from a monaural sound source ).

Background technique

A typical surround sound system uses some simple delay circuits and phase filters to mix the left and right channel data to simulate a stereo sound effect. However, this method usually confuses the data of the left and right channels due to the mixing process, resulting in distortion of the original data. Again, the above-mentioned typical surround sound system cannot produce stereophonic effects from four directions of a listener's front, rear, up and down, especially when the listener is not located in a so-called "sweet spot". Therefore, the above-mentioned system Filters need to be installed at the position of the front and rear sound, and then an electrical system is provided to allow the listener to hear the front sound through the front filter when the listener turns his head 180 degrees from left to right, and turn from right to right. When left, the sound behind can be heard through the filter behind. According to the system data derived from the numerical analysis, the scaler is used on the output and input of the filter to adjust the range and position of the sound source. To eliminate such as ITD and two hearing sound density differences (11D) (described later). However, this will require a large amount of circuit components and filter power to provide on-the-spot sound effects. Therefore, the HRTF technology commonly used to produce stereo sound effects is developed The above-mentioned HRTF provides a database, and the database is built in to measure the frequency response characteristics of the ear and the brain when the sound emitted at each given position is transmitted to the two ears in a 360-degree space. Using a digital signal processor ( DSP) calculates the above-mentioned frequency multi-response characteristics to calculate the sound wave with stereo sound effect. When the sound wave is transmitted to the ear, the brain can feel the 360-degree reality and produce the effect of stereo sound. Figure 1 shows A typical 360-degree HRTF measurement.As shown in Figure 1, a dummy head 12 is placed at the center of a circle 10 and the circle 10 is divided into 360 sections, and each section specifies an azimuth control parameter (360-degree control) and Represents a non-repetitive positioning point. In this way, when the position of a light source is smoothly transferred from one section to the next, the dummy head 12 can perceive the continuous movement of the light source position. According to the above HRTF measurement, it is assumed that position 0 is aligned with head 12 The left ear of the head is represented by the number 14. The position 90 is directly in front of the head 12 and is represented by the number 16. Similarly, the position 180 is aligned with the right ear of the head 12 and is represented by the number 18, and the position 270 is directly behind the head 12 and is represented by the number 18. Number 20 represents. Because the azimuth position parameters are configured around the circle 10, therefore, positions 0 and 360 will overlap at the number 14. A range parameter is used to represent the sound range or distance of the sound source. Take the center point of head 12 is the origin (position 0), and is divided into 360 circles from the origin to circle 10 to represent the sound source distance. Among them, positions 0 to 19 are the range of the head 12, as shown in the figure number 22. The remaining positions 20 to 359 are numbered in the figure 24. The above-mentioned position 359 is the limit of the hearing range. Of course, the above-mentioned limit can be adjusted according to actual needs.

The aforementioned configuration must be built in an echoic chamber, and binaural measurements are performed under each sound source in a 360-degree (segment) different coordinate space to record 20HZ to 20KHZ at a sampling rate of 48 or 44.IKHZ sound waves. Because the standard time difference between the sound waves transmitted to the two ears is about the time for the sound waves to travel 20 centimeters (range 0-19), resulting in the sound time difference between the two ears (ITD). In addition, different people's heads, shoulders, and arms will absorb different sound wave energy. At this time, the amplitudes of the loud and small sounds heard by the two ears are different, resulting in a difference in sound density between the two ears (11D). In this way, a single sound source is used to measure and record the HRTF of the left and right ears respectively, so as to complete the establishment of the HRTF database. But this way is well known to those who are familiar with this technology, the sound wave will have a reflection effect when it hits an object, and the sound waves of different wavelengths changed by the human body, shoulder, head and outer ear will produce diffraction in the outer ear shell ( diffractive) phenomenon, so that the frequency response of the sound felt by the drum is also different. These changes recorded in the HRTF database not only vary with the azimuth, elevation, range and frequency of the point sound source, but also vary from person to person. Therefore, it is usually quite laborious to establish this kind of HRTF database. Generally speaking, adding It takes several months to complete the adjustment of a complete set of HRTF parameters.

Contents of the invention

In view of this, an object of the present invention is to provide a method for generating a stereo sound effect, which uses a Head Response Transformation algorithm (HRTF) to synthesize a binaural sound effect from a monaural sound source.

The generation method of stereophonic effect of the present invention comprises the following steps:

Measurements are taken on both ears to obtain the transfer function for both ears;

Recording a set of HRTF coefficients of one ear through a monaural HRTF measurement operation to establish a monaural head correspondence transformation algorithm (HRTF) database;

According to the conversion function of the two ears to calculate the above-mentioned interaural sound time difference (ITD) compensation curve; and

According to the above two established monaural HRTF databases and ITD compensation curves, an externally input monaural signal is converted into a stereo sound effect signal.

The method for generating stereo sound effect further includes the following steps of adjusting the ITD module of the stereo sound effect signal according to the ITD compensation curve. The above method for generating stereo sound effect further includes providing an adjustment tool for a user to set the head shadow parameters to the expected surround effect of stereo sound effect.

Another stereophonic production method pointed out by the present invention is to measure the two ears to obtain the conversion function of the two ears;

Recording a set of HRTF coefficients of one ear through a monaural HRTF measurement operation to establish a monaural head correspondence transformation algorithm (HRTF) database;

Calculate the sound time difference (ITD) compensation curve between the above-mentioned two ears according to the above-mentioned conversion function of the above-mentioned two ears;

Separate an external input monaural signal into a near-end binaural signal and a far-end binaural signal according to the established HRTF database;

adjusting an ITD module of the above-mentioned remote binaural signal according to the established HRTF database and ITD compensation curve;

An adjustment tool is provided for a user to set the head shadow parameters of both the near-end binaural signal and the far-end binaural signal with the adjusted ITD module to a desired stereo sound surround effect.

The present invention can use HRTF technology to synthesize a binaural sound effect from a monaural sound source, thereby making stereo sound effect setting more convenient and saving adjustment time.

Description of drawings

Figure 1 shows a typical HRTF measurement map with 360 degrees;

Fig. 2 shows a single ear HRTF measurement chart of the present invention;

Fig. 3 shows a two-channel synthesizing structure diagram according to the present invention;

Fig. 4 shows a graph of ITD compensation used in the ITD oscillator of Fig. 3 according to the present invention;

FIG. 5 is a flowchart of the operation of FIG. 3 according to the present invention.

Detailed ways

Throughout, similar element functions are denoted by the same element number.

Figure 2 shows a single-ear HRTF measurement diagram of the present invention. In Fig. 2, the above single-ear HRTF measurement only records the measurement data of the left ear or the right ear (a set of filter numbers), which is different from the typical HRTF measurement, which must record the measurement data of the left ear and the right ear separately. As shown in FIG. 2, a loudspeaker 21 is placed 1.4 meters away from the left ear of the head 22 (as a receiving microphone) for transmitting sound to the left ear. Since the human face is roughly symmetrical, when performing HRTF measurement, it is possible to measure only the right half face or the left half face, at each measurement point (for example, at the preset azimuth angle θ _AC and elevation angle θ At point A of _AB ) record the sound broadcast from the loudspeaker sampled by the microphone at 44.1kHZ, record a set of filter numbers (impulse response) in the frequency domain and obtain the transfer function between them ) (conversion from the frequency domain to the time domain), the HRTF database can be completed. Also, a time equalizer (not shown) can be used to remove the effects of the measuring instrument, such as the frequency response of the loudspeaker 21, head 22 and receiving microphone (ear on the measuring side) itself. In this way, the measurement results stored in the HRTF database will be conversion functions independent of the measurement instrument itself including the header. In addition, it is actually difficult to directly implement the above database, because it must be possible with a very long FIR filter, usually hundreds of taps or more, and we have to do further Simplification. Since the phase difference between the transfer function to the left ear and the transfer function to the right ear can be included in the ITD module to compensate, we perform a minimum phase shift on the transfer functions for both ears, and then observe the converted results It is found that most of the energy is concentrated in the front 32 taps. In this way, as long as the first 32 orders of the conversion function are stored in the conversion function generated by the measurement data through the minimum phase conversion, the 3-D stereoscopic effect can be effectively simulated. The aforementioned HRTF database including the number of FIR filters of the first 32 orders can be implemented simply by using a 32-order FIR filter. However, considering that each person’s head shape is different in practical applications, in order to remove the influence of the head 22 on the conversion function, refer to Figure 3, whether the sound to the left ear or the right ear will pass through the head shadow effect The compensation filters 32, 34 are used to compensate the different perceptions of the 3-D stereo sound effects of different head shapes. In addition, according to acoustic psychology, the delay from the near-end loudspeaker to the left and right ears is so small that it can be ignored, but the time difference between the far-end loudspeaker and the two ears cannot be ignored, so an ITD filter needs to be used to compensate. In this way, individual and different 3-D stereo sound effects can be achieved by using a very simple filter structure, so as to save calculation costs and time. FIG. 4 shows a graph of ITD compensation used in the ITD filter of FIG. 3 according to the present invention. In FIG. 4 , the above compensation curve is the ITD module when the elevation angle is 0 degrees (that is, the C axis in FIG. 2 ). As shown in Figure 4, at the same measurement position (azimuth angle) in the original measurement database, the delay part of the conversion function measured by the left ear and the conversion function measured by the right ear is taken out for cross-correlation calculation (crosscorrelation) (can be obtained by multiplying the Gaussian function by the sine function) to find the maximum delay value, which is the ITD reference value we want to compensate. Next, the ITD compensation curve derived in Fig. 4 was placed between the monaural HRTF database and the far-end IIR filter with the ITD filter shown in Fig. 3 . Also, based on the above-mentioned different head shapes of each person, in addition to establishing a common 32-order FIR filter to simulate stereo sound effects, an adjustment tool is also provided for users to set the head of the near-end and far-end shadow effect wave filters. Inner shadow parameters, which provide a way to adjust individual differences to achieve stereo surround effects. This adjustment tool provides two parameter setting devices to adjust the pole and zero values of each IIR filter to a level that allows the user to clearly feel the stereo effect.

In summary, the operation flow of the structure in FIG. 3 is shown in FIG. 5 . In Fig. 5, the operation flow includes the following steps: establishing a single-ear HRTF database and an ITD compensation curve (S1); and performing ITD adjustment and shadowing effect adjustment according to the established single-ear HRTF database and ITD compensation curve (S2). As shown in FIG. 5 , the above-mentioned establishment of a monaural HRTF database includes 32-order FIR filter coefficients and a 32-order FIR filter is used to implement the monaural HRTF database. Although 32 levels are used as an example for illustration here, in practice, the required level for configuring the above database can be determined according to actual needs, and is not limited to 32 levels. The slope of the established ITD compensation curve will present a curve that is close to a proportional constant (as shown in Figure 4). In this way, the user can adjust the ITD module of the present invention and the far and near end head shadow effect filter (IIR filter) , easily find the setting that you think the 3-D effect is the most obvious. Therefore, through the stereo sound effect generation method of the present invention, it is not necessary to perform HRTF measurement on individual differences and/or change the number of filters stored in the database to generate stereo sound effects that satisfy users.

Although the present invention has been disclosed as above with preferred embodiments, it is not intended to limit the present invention. Anyone familiar with this art can make some changes and modifications without departing from the spirit and scope of the present invention. Therefore, this The scope of protection of an invention shall be defined by the scope of the claims.

Claims (8)

1. A method for producing stereophonic effects is characterized in that it may further comprise the steps: Measurements are taken on both ears to obtain the transfer function for both ears; Recording a set of HRTF coefficients of one ear through a monaural HRTF measurement operation to establish a monaural head correspondence transformation algorithm (HRTF) database; According to the conversion function of the two ears to calculate the above-mentioned interaural sound time difference (ITD) compensation curve; and According to the above two established monaural HRTF databases and ITD compensation curves, an externally input monaural signal is converted into a stereo sound effect signal. 2. the generation method of stereophonic effect as claimed in claim 1 is characterized in that: this builds up a monaural head correspondence conversion algorithm (HRTF) data bank to comprise finite impulse response (FIR) filter coefficient, and above-mentioned finite impulse Response (FIR) filter coefficients are implemented as a FIR filter. 3. The method for generating stereo sound effect according to claim 1, further comprising a step of adjusting the ITD module of the stereo sound effect signal according to the ITD compensation curve. 4. The method for generating stereo sound effects as claimed in claim 1, further comprising a step of providing an adjustment tool for a user to set the head shadow parameters to the expected surround effect of stereo sound effects. 5. The method for generating stereo sound effects as claimed in claim 4, wherein the head shadow parameters include infinite impulse responses (IIRs). 6. A method for producing stereophonic effects, characterized in that: comprise the following steps: Measurements are taken on both ears to obtain the transfer function for both ears; Recording a set of HRTF coefficients of one ear through a monaural HRTF measurement operation to establish a monaural head correspondence transformation algorithm (HRTF) database; Calculate the sound time difference (ITD) compensation curve between the above-mentioned two ears according to the above-mentioned conversion function of the above-mentioned two ears; Separate an external input monaural signal into a near-end binaural signal and a far-end binaural signal according to the established HRTF database; adjusting an ITD module of the above-mentioned remote binaural signal according to the established HRTF database and ITD compensation curve; An adjustment tool is provided for a user to set the head shadow parameters of both the near-end binaural signal and the far-end binaural signal with the adjusted ITD module to a desired stereo sound surround effect. 7. the generation method of stereophonic effect as claimed in claim 6 is characterized in that: this establishes a monaural head corresponding transformation algorithm HRTF database to comprise finite impulse response (FIR) filter coefficient, and above-mentioned finite impulse response ( FIR) filter coefficients are implemented as a FIR filter. 8. The method for generating stereo sound effects as claimed in claim 6, wherein the head shadow parameters include infinite impulse responses (IIRs).

Images