JP2011160179A - Voice processor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a virtual surround system capable of suppressing increase in an operation amount when an upper virtual sound source is added. <P>SOLUTION: The upper virtual sound source is reproduced by a speaker arranged only at a front position while suppressing the operation amount for voice processing by processing a lower-frequency component of an upper speaker signal and a horizontal speaker signal extracted by a low-pass filter, by an FIR filter, and processing a high-frequency component of the upper speaker signal extracted by a high-pass filter, by an IIR filter. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a sound processing apparatus that creates a realistic sound field using a plurality of speakers, and more particularly to a virtual surround system that creates a virtual sound source using speakers arranged only at the front.

  In recent years, with the widespread use of DVD (Digital Versatile Disc) and BD (Blu-ray Disc), surround systems that provide high-realistic sound by outputting sound from speakers placed in front of and in front of the listener Demand is increasing.

  In the BD standard, in addition to a speaker arrangement that is assumed to be arranged at the same height as the listener's ear, a speaker arrangement that is assumed to be arranged above the listener is prepared. It becomes possible to create a sense of reality. Therefore, in content that requires a sense of realism in the height direction, a channel signal for the upper speaker may be incorporated in the future.

  In addition, there is a virtual surround system that reproduces all speakers included in the surround system as virtual speakers using only front speakers arranged in front of the listener. In the system, generally, a transfer characteristic from a desired speaker position to a listener's ear called HRTF (Head-Related Transfer Function) is arranged in front of the front speaker. The localization of the virtual speaker is realized by reproducing at the listener's ear using a response) filter (Patent Documents 1 and 2).

  FIG. 3 is a configuration diagram when one virtual speaker 304 is created behind the listener by using two front speakers (left speaker 302 and right speaker 303) arranged in front of the listener 301. FIG. 4 shows a virtual processing unit 305 for creating a virtual speaker 304. Localization of the virtual speaker 304 behind the listener is realized by reproducing the HRTF from the virtual speaker 304 to the listener's ears with the listener's ears by using the left and right FIR filters 401 and 402 in front of the left and right front speakers 302 and 303. it can. In particular, the FIR filters 401 and 402 faithfully reproduce the mid-high frequency peaks and dip characteristics that appear in the HRTF from the virtual sound source to the listener's both ears, thereby increasing the localization accuracy of the virtual sound source.

JP-A-8-265899 JP 2008-154082 A

  However, when an attempt is made to realize a bar channel surround system using the FIR filter for the upper virtual sound source in addition to the horizontal direction, there is a problem that the amount of calculation increases because the processing increases.

  Accordingly, an object of the present invention is to provide a speech processing apparatus that can suppress an increase in the amount of calculation even when an upper virtual sound source is added in order to solve the above-described problem.

  In order to achieve the above object, an audio processing apparatus of the present invention includes a first delay unit that adjusts an output timing of a horizontal speaker signal, a low-pass filter that inputs an upper speaker signal and outputs a low-frequency component, Second delay means and third delay means for adjusting the output timing of the signal from the low-pass filter, and first addition for adding the signal from the first delay means and the signal from the second delay means Means, a second adding means for adding the signal from the first delay means and the signal from the third delay means, a high-pass filter for inputting an upper speaker signal and outputting a high-frequency component, Fourth delay means and fifth delay means for adjusting the output timing of a signal from the high-pass filter, a first filter for inputting a signal from the first adder means, and the second adder A second filter that receives a signal from the fourth delay means, a third filter that receives a signal from the fourth delay means, a fourth filter that receives a signal from the fifth delay means, and the second filter. A third adding means for adding the signal from the first filter and the signal from the third filter; and a fourth adding means for adding the signal from the second filter and the signal from the fourth filter. It is characterized by comprising.

  According to said structure, the amount of calculations can be reduced in the audio | voice processing apparatus corresponding to the virtual reproduction of a horizontal direction and an upper speaker.

Speaker configuration diagram of the speech processing apparatus according to the first embodiment. Virtual processing unit block diagram of the speech processing apparatus according to the first embodiment Speaker configuration diagram of audio processing apparatus according to prior art Block diagram of virtual processing unit of speech processing apparatus according to prior art

Hereinafter, a virtual surround system will be described as an example of an embodiment of a sound processing apparatus of the present invention with reference to the drawings.
(Embodiment 1)
FIG. 1A shows a side view of the virtual surround system according to the first embodiment, and FIG. 1B shows an upper side view.

  The virtual surround system includes a left speaker 102, a right speaker 103, a horizontal virtual speaker 104, an upper virtual speaker 105, and a virtual processing unit 106.

  In FIG. 1, front speakers (left speaker 102 and right speaker 103) are arranged on the front left and right of the listener 101, and in the horizontal direction (elevation angle 0 °) behind the listener using sound waves radiated from both speakers. The horizontal virtual speaker 104 intended for localization and the upper virtual speaker 105 intended for localization with the same azimuth angle and upward elevation angle as the virtual speaker 104 are created.

  FIG. 2 is a block diagram of the virtual processing unit of the virtual surround system according to the first embodiment.

  The virtual processing unit 106 includes FIR filters 201 and 202 for horizontal virtual speakers, IIR filters 203 and 204 for left and right front speakers, delay processing units 205 to 209, LPF (Low Pass Filter) 210, HPF (High). (Pass Filter: high-pass filter) 211.

  The horizontal virtual speaker FIR filters 201 and 202 each convert an input signal into a desired transfer characteristic (including delay amount control).

  The left and right front speaker IIR (Infinite Impulse Response) filters 203 and 204 each convert an input signal into a desired transfer characteristic (not including delay amount control).

  Each of the delay processing units 205 to 209 adjusts (delays control) the output timing of the input signal.

  The LPF 210 extracts a low frequency component of the input signal.

  The HPF 211 extracts a high frequency component of the input signal.

  The virtual processing unit 106 has different processing methods for the horizontal direction signal and the upward signal. That is, the horizontal direction input signal is converted into a signal localized in the horizontal direction virtual speaker 104 by being processed by the horizontal direction virtual speaker FIR filters 201 and 202 as in the conventional technique.

  Here, in general, humans are vague in the direction perception of the elevation direction (height direction) compared to the horizontal direction, and particularly from the viewpoint of HRTF, the low frequency component has a small characteristic difference due to the difference in the elevation direction. Therefore, even when the low-frequency component of the upper speaker signal is regarded as a horizontal speaker signal and virtual processing is performed, it is possible to suppress the deterioration of the localization accuracy of the upper speaker.

  Next, the operation of the virtual processing unit 106 will be described.

  First, the upper input signal is subjected to band division by the LPF 210 and the HPF 211.

  Next, the low frequency components output from the LPF 210 are processed by the horizontal virtual speaker FIR filters 201 and 202, and the high frequency components output from the LPF 211 are processed by the left and right front speaker IIR filters 203 and 204. By adding each signal of the left and right front speakers, the signal is converted into a signal localized at the upper virtual speaker 105.

  The IIR filter has a feature that a desired frequency characteristic (in the present invention, a peak or dip in a specific frequency band of the HRTF from the virtual speaker to the listener's ear) can be realized with a small amount of calculation compared to the FIR filter. In this embodiment, by using the IIR filter for the upper virtual speaker, it is possible to realize the localization of the upper virtual speaker 105 in addition to the horizontal virtual speaker 104 of FIG.

  By the way, the FIR filters 201 and 202 and the IIR filters 203 and 204 have different time delays generated by processing (the delay amount cannot be controlled by the IIR filter). The desired delay amount to be made is different. Therefore, in order to adjust the output timing after the virtual processing of the upper direction input signal in the horizontal direction, the delay processing units 205 to 209 are arranged in the pre-filter processing stage.

  3 expands the rear virtual speaker arrangement shown in FIG. 3 to the front, the horizontal direction, and the like, and performs virtual processing shown in FIG. 4 to localize the virtual speakers in the horizontal and upper directions so as to wrap the listener. The system can be realized while suppressing an increase in calculation amount.

  In this embodiment, the low-frequency component of the horizontal direction signal and the upward signal are processed by the FIR filter, and the high-frequency component of the upward signal is processed by the IIR filter. After combining the parameters of the FIR filter, IIR filter and delay processing unit, it is also possible to process the low frequency component of the horizontal signal and the upward signal with the IIR filter and the high frequency component of the upward signal with the FIR filter. is there. If both processing configurations are prepared, it is possible to perform a switching operation in which an input signal desired to be localized more accurately in accordance with the content or scene is virtually processed with an FIR filter and the other is virtually processed with an IIR filter.

  The present invention relates to a surround system that creates a realistic sound field using a plurality of speakers, and more particularly to virtual playback that creates a virtual sound source using speakers arranged only at the front, and an upper virtual sound source. It is possible to provide a virtual surround system that can suppress an increase in the amount of computation when adding the. Therefore, the present invention can be applied to a home theater system including a plurality of speakers.

101 listener 102 left speaker 103 right speaker 104 horizontal virtual speaker 105 upper virtual speaker 106 virtual processing unit 201, 202 FIR filter 203, 204 IIR filter 205, 206, 207, 208, 209 delay processing unit 210 LPF
211 HPF

Claims (2)

First delay means for adjusting the output timing of the horizontal speaker signal;
A low-pass filter that inputs the upper speaker signal and outputs a low-frequency component;
Second delay means and third delay means for adjusting the output timing of the signal from the low-pass filter;
First addition means for adding the signal from the first delay means and the signal from the second delay means;
Second addition means for adding the signal from the first delay means and the signal from the third delay means;
A high-pass filter that inputs an upper speaker signal and outputs a high-frequency component;
Fourth delay means and fifth delay means for adjusting the output timing of the signal from the high-pass filter;
A first filter for inputting a signal from the first adding means;
A second filter for inputting a signal from the second adding means;
A third filter for inputting a signal from the fourth delay means;
A fourth filter for inputting a signal from the fifth delay means;
Third addition means for adding the signal from the first filter and the signal from the third filter;
An audio processing apparatus comprising: a fourth adding unit that adds a signal from the second filter and a signal from the fourth filter. The audio processing apparatus according to claim 1, wherein the first filter and the second filter are FIR filters, respectively, and the third filter and the fourth filter are IIR filters, respectively.

Images