WO2010094219A1

WO2010094219A1 - Method and device for processing and reproducing speech signals

Info

Publication number: WO2010094219A1
Application number: PCT/CN2010/070491
Authority: WO
Inventors: 梁丽燕; 刘智辉
Original assignee: 华为终端有限公司
Priority date: 2009-02-19
Filing date: 2010-02-03
Publication date: 2010-08-26
Also published as: CN101510988A; CN101510988B

Abstract

A method and device for processing and reproducing speech signals are provided, wherein said processing method includes: according to location information set for conference halls participating in the conference, determining the location information of the conference hall having the maximum energy at each time instant and in each frequency band among the maximum N-party conference halls; transmitting the mixed speech signals of the maximum N-party conference halls and the location information of the conference hall having the maximum energy at each time instant and in each frequency band to terminals in the conference halls participating in the conference. The reproducing method includes: obtaining the mixed speech signals and the location information of the conference hall having the maximum energy in each frequency band; according to the correspondences between auditory spatial parameters of a reproduction device and location information, obtaining the auditory spatial parameters of the reproduction device corresponding to the location information of the conference hall having the maximum energy in each frequency band; adjusting said mixed speech signals by using the auditory spatial parameters of the reproduction device; and reproducing the mixed speech signals after being adjusted. According to the embodiment of present invention, the spatially auditory effect of the video conference is improved.

Description

Processing and playing method and device for voice signal

This application claims priority to Chinese Patent Application No. 200910005681.X, filed on February 19, 2009, entitled "Processing, Playback, and Device for Voice Signals", the entire contents of which are hereby incorporated by reference. The citations are incorporated herein by reference.

Technical field

The present invention relates to the field of video communication technologies, and in particular, to a method and apparatus for processing and playing a voice signal.

Background technique

In the video communication system, each conference site participating in the conference encodes the local voice signal and image signal and sends it to the MCU (Multipoint Control Unit), and the MCU processes the received voice signal and image signal, and processes the signal. The subsequent voice signal and image signal are sent to each venue terminal, and the conference field plays the decoded voice signal and the image signal, thereby realizing video communication. The MCU calculates the envelope of the voice signal of each site after the speech signal is processed, and compares the envelopes of the voice signal to the N sites with the largest envelope as the largest N-party venue, and then maximizes The voice signal of the N-party site is mixed and sent to other sites outside the largest N-party site in the conference. The voice signal received by the largest N-party site is the voice signal of the largest N-1 party site other than the site where it is located. Mix signal. Therefore, after the conference site decodes the received mix signal, the other N-party sites can hear the voice of the largest N-party site, and the largest N-party can hear the other N-1. The voice of the party venue.

However, the inventors found in the prior art that in the prior art, when the MCU mixes the speech signals of the largest N-party venue, only the speech signals of the largest N-party venue are linearly superimposed. When the maximum N-party site is spoken at the same time, the sound played by the output devices of each participating site is the mixed and overlapping voice of each of the largest N-party sites, so that the users participating in the conference cannot hear each of the largest N-party. The sound of the venue, which affects the audiovisual effect of the video conference. Summary of the invention

Embodiments of the present invention provide a method and apparatus for mixing and playing a voice signal to improve the spatial hearing effect of the video conference.

The embodiment of the invention discloses a method for processing a voice signal, comprising: determining, at a maximum N-party venue, each time in the mixed signal according to the orientation information set for the meeting place participating in the conference Azimuth information of the site with the largest energy in the frequency band; the sounding signal of the largest N-party site and the orientation information of the site with the largest energy in each frequency band at each time are sent to the conference terminal participating in the conference.

The embodiment of the invention further discloses a method for playing a voice signal, comprising: acquiring a mixing signal of a maximum N-party venue and a position information of a site with the largest energy in each frequency band at each time; according to the auditory space parameter of the playing device Corresponding relationship between the orientation information, obtaining an auditory spatial parameter of the playback device corresponding to the orientation information of the site with the largest energy at each frequency band at each time; adjusting the mix by using the auditory spatial parameter of the playback device Signal, and play the adjusted mix signal.

The embodiment of the invention further discloses a processing device for a voice signal, comprising: an orientation determining unit, configured to determine each moment in the mixed signal in the largest N-party venue according to the orientation information set for the meeting place participating in the conference The orientation information of the site with the largest energy in each frequency band; the sending unit, configured to send the sound mixing signal of the largest N-party meeting site and the position information of the site with the largest energy in each frequency band at each time to the conference terminal participating in the meeting .

The embodiment of the invention further discloses a playback device for a voice signal, comprising: an acquisition unit, configured to acquire a sound mixing signal of a maximum N-party venue and a position information of a site with the largest energy in each frequency band at each time; a unit, configured to obtain, according to a correspondence between the auditory spatial parameter and the azimuth information of the playback device, an auditory spatial parameter of the playback device corresponding to the orientation information of the site with the largest energy in each frequency band at each time; And for adjusting the mixing signal by using an auditory space parameter of the playing device, so as to play the adjusted mixing signal.

It can be seen from the foregoing embodiment that, when processing the voice signal, the orientation information is set in advance for all the sites participating in the conference, and in the largest N-party conference site, the orientation of the site with the largest energy in each frequency band at each moment is determined. Information, the orientation information is sent together with the mixing signal of the largest N-party venue. When playing the voice signal, according to the corresponding relationship between the received orientation information and the spatial parameters of the playing device, the spatial parameters of each playing device at the playing end are obtained, and the spatial parameters of the playing device are used to adjust the mixing signal. When the adjusted mixing signal is played, the auditory space of each mixing site can be reconstructed at the venue, so that the sound of the largest N-party venue has a spatial stereoscopic feeling during playback, and the user can clearly understand each of the largest N-party venues. The sound of the user adds to the user's experience of the spot.

DRAWINGS

+ 1 is a flowchart of a method for processing a voice signal according to Embodiment 1 of the present invention; FIG. 2-a is a schematic diagram of a position of 10 conference venues;

Figure 2-b is a schematic diagram of the orientation of four sites in a multi-screen;

Figure 3-a shows the orientation of the four largest 4-party venues;

Figure 3-b is a schematic diagram of the orientation of four sites in a multi-screen;

Figure 4 shows the setting method of the orientation when the number of multi-screens is 16, and the number of orientations is 4;

Figure 5 is a schematic diagram of processing of a voice signal in the present invention;

FIG. 6 is a structural diagram of a method for processing a voice signal according to Embodiment 2 of the present invention; FIG. 7 is a flowchart of a method for playing a voice signal according to Embodiment 3 of the present invention; Schematic diagram of adjusting the auditory spatial parameters of the playback device;

FIG. 9 is a structural diagram of a device for playing a voice signal according to Embodiment 4 of the present invention.

detailed description

The above described objects, features, and advantages of the present invention will become more apparent from the aspects of the invention. Embodiment 1 Referring to FIG. 1 , FIG. 1 is a flowchart of a method for processing a voice signal according to the present invention, where the method includes the following steps:

Step 101: Determine, according to the orientation information set for the conference venue participating in the conference, the location information of the site with the largest energy in each frequency band at each moment in the mixed signal in the maximum N-party conference site;

In the above steps, the voice signal of the largest N-party site is first time-frequency transformed, the voice signal in the time domain is converted into a voice signal in the frequency domain, and then the energy value in each frequency band at each time is calculated, and each time is obtained. At the moment, the site with the highest energy in each frequency band is finally determined according to the orientation information set for the meeting place participating in the meeting, and the orientation information of the site with the largest energy in each frequency band is determined. The location information of the site with the largest energy in the largest N-party site in each frequency band at each time can be determined by two methods.

A method for determining the method is as follows: According to the order of joining the conference sites participating in the conference, the orientation of the conference site is set in advance. When comparing the energy of each band in the speech signal by comparing the largest N-party venue After obtaining the site with the largest energy in the largest N-party site in each frequency band at each time, it is determined whether the site with the largest energy is in the multi-screen, and if so, the orientation information of the site with the largest energy is set to be more The screen orientation information, if not, sets the orientation information of the site with the largest energy to the preset orientation information. For example, in a videoconferencing system, there are ten venues participating in the conference. The first conference site number is 1, the second conference site number is 2, and so on. The tenth conference site number is 10. According to the order of joining the ten venues, the orientation of field 1-3 is set to the upper left, the orientation of field 4-6 is set to the upper right, and the orientation of field 7-8 is set to the lower left, which will be 9-10. The orientation is set to the lower right, please refer to Figure 2-a. Figure 2-a shows the orientation of the 10 joining venues. The site 1-4 is the largest 4-party site, and in a certain frequency band at a certain time, the site 1 is the site with the largest energy in the largest 4-party site, and it is determined whether the site 1 is in the multi-screen, when the site 1 In multi-screen, set the orientation information of field 1 in multi-screen to the orientation information of site 1. For example, site 1 is at the bottom right of the multi-screen, see Figure 2-b, Figure 2-b is multi-screen. The orientation information of the four sites is the right lower part of the site. When the site 1 is not in the multi-screen, the orientation information of the site is obtained. The orientation information of the site 1 is the upper left.

The other method is as follows: After determining the maximum N-party site, the orientation of the largest N-party site is set in advance according to the order of joining the largest N-party site, and the orientation information of the largest N-party site is obtained. When comparing the energy value of each frequency band of the voice signal of the largest N-party site at each time to obtain the site with the largest energy in the largest N-party site in each frequency band at each time, it is determined whether the site with the largest energy is at most In the picture, when the site with the largest energy is in the multi-screen, the orientation information of the site with the largest energy is set as the orientation information of the site with the largest energy in the multi-screen, and if not, the site with the largest energy. The orientation information is set to the orientation information of the preset maximum N-party venue. Take the video communication between the above ten sites as an example. The venues 1-4 are the maximum 4-party venues. According to the order of joining the venues 1-4, the orientation of the field 1 is set to the upper left, and the orientation of the field 2 is set. For the upper right, set the orientation of field 3 to the lower left, and set the orientation of field 4 to the lower right. See Figure 3-a. Figure 3-a shows the orientation of the four largest 4-party venues. When comparing the energy value, it is known that when the site 1 is the site with the largest energy in the largest 4-party site in a certain frequency band at a certain time, it is determined whether the site 1 is in the multi-screen, and when the site 1 is in the multi-screen, The orientation information of the site 1 in the multi-screen is set to the orientation information of the site 1. For example, the site 1 is at the lower right of the multi-screen. Please refer to Figure 3-b. Figure 3-b shows the orientation of the four sites in the multi-screen. , the location information of field 1 is When the site 1 is not in the multi-screen, it can be obtained according to the preset position of the maximum 4-party site. The orientation information of the site 1 is the upper left.

In the above two determination methods, when the orientation of the site with the largest energy changes in the multi-picture, the orientation information of the site with the largest energy also changes correspondingly with the change of the orientation.

For example, the video communication between the above 10 sites is used as an example. The conference site 1-4 is the maximum 4-party venue. According to the order of joining the conference sites 1-4, the orientation of the field 1 is set to the upper left, and the orientation of the field 2 is Set to the upper right, set the orientation of field 3 to the lower left, and set the orientation of field 4 to the lower right. Moreover, in a certain frequency band at a certain time, the site 1 is the site with the largest energy in the largest 4-party site, and in the multi-screen, the orientation information of the field 1 is the orientation information of the site 1 in the multi-screen, assuming The orientation of the site 1 in the multi-screen is the upper left, and the orientation information of the field 1 is the upper left. When the orientation of the site 1 in the multi-screen is switched to the upper right, the site 1 is still in the maximum 4-party venue. When the site with the highest energy is used, the orientation information of field 1 changes accordingly to the upper right.

It should be noted that, in this embodiment, the method for setting the orientation information of the site with the largest energy in the largest N-party site is not limited, and the orientation information is not limited to the four directions of the upper left, the upper right, the lower left, and the lower right.

When the number of pictures of a multi-picture is larger than the number of azimuth, the site in the multi-picture cannot completely correspond to any one of the positions. For example, if the number of multi-pictures is 16, and the number of azimuths is 4, the site in the multi-picture cannot correspond to any one of the positions, and the most similar position is taken for the site in the multi-picture. Please refer to Figure 4, 4 is a setting method in which the number of multi-pictures is 16, and the number of orientations is 4, and the orientation of the venue 7 in the figure is set to the upper right according to the approximation principle.

Step 102: Send the mixed signal of the largest N-party venue and the orientation information of the site with the highest energy in each frequency band at each time.

In the foregoing step 102, the sound mixing signal of the largest N-party site and the orientation information of the site with the largest energy in each frequency band at each time are first encoded, respectively, and the mixed code stream and the position information stream are respectively obtained, and then the sound is mixed. The code stream and the azimuth information code stream are sent to the site terminal participating in the conference; or, only the mixed signal of the largest N-party site may be encoded to obtain a mixed code stream, and then the mixed code stream and each time of each time The location information of the site with the largest energy in the frequency band is sent to the site terminal participating in the conference. For example, if the destination site belongs to the largest N-party site, the mix signal sent to the site is the mix signal of the largest N-1 site other than the site. Wherein, when the orientation information of the site having the largest energy is encoded, the encoding is performed in an encoding manner different from the mixing signal. For example, when the mixing signal is encoded according to the conventional encoding protocol G.722, the orientation information of the site with the largest energy in the largest N-party site can be encoded by Huffman. Please refer to FIG. 5. FIG. 5 is a schematic diagram of processing of a voice signal according to the present invention.

When the orientation information of the site with the largest energy in each frequency band is transmitted at each time, another implementation manner is:

The site number of the site with the largest energy in the largest N-party site is transmitted together with the location information of the largest N-party site. The receiver determines the maximum energy of the largest N-party site from the location information of the largest N-party site based on the site number. Location information of the venue. Embodiment 2 Corresponding to the processing method of a voice signal, the embodiment of the present invention further provides a processing device for a voice signal. Referring to FIG. 6, FIG. 6 is a structural diagram of a processing apparatus for a voice signal according to the present invention. The apparatus includes an orientation determining unit 601 and a transmitting unit 602. The internal structure and connection relationship will be further described below in conjunction with the working principle of the device.

The position determining unit 601 is configured to determine, according to the orientation information set for the meeting place of the meeting, the orientation information of the site with the largest energy in each frequency band at each time in the largest N-party meeting place;

The sending unit 602 is configured to send the sound mixing signal of the largest N-party venue and the orientation information of the site with the largest energy in each frequency band at each time.

The orientation determining unit 601 may include: a first orientation determining unit 603, configured to pre-set an orientation for the conference site participating in the conference according to the order of joining, to obtain preset orientation information; and comparing unit 604, for comparing The maximum value of the energy value of the voice signal of each of the N-party sites in each frequency band is obtained, and the first setting unit 605 is configured to: when the site with the largest energy is not in the multi-picture, according to The preset orientation information sets the orientation information of the site with the largest energy; the second setting unit 606 is configured to set the orientation information of the site with the largest energy according to the multi-screen orientation information when the site with the largest energy is in the multi-screen.

The orientation determining unit 601 may further include: a second orientation presetting unit, configured to pre-set the orientation for the largest N-party venue according to the order of joining, and obtain preset orientation information of the largest N-party venue; a comparison unit, configured to compare energy values of each frequency band of the voice signal of the largest N-party site at each time, to obtain a site with the largest energy in each frequency band at each time; and a third setting unit, configured to use the maximum energy When the site is not in the multi-screen, the orientation information of the site with the largest energy is set according to the preset orientation information. The fourth setting unit is configured to set the maximum energy according to the multi-screen orientation information when the site with the largest energy is in the multi-screen. Location information of the venue.

The sending unit 602 may include: a first sending unit 607 and/or a second sending unit 608, where the first sending unit 607 is configured to use the mixed signal and the maximum energy in each frequency band at each time. The orientation information is encoded, and the mixed code stream and the position information stream are respectively obtained, and the mixed code stream and the position information code stream are sent to the conference terminal participating in the conference;

The second sending unit 608 is configured to encode the mixed signal to obtain a mixed code stream, and send the mixed code stream and the position information of the site with the largest energy in each frequency band at each time to participate. The venue terminal of the conference. Embodiment 3 Referring to FIG. 7, FIG. 7 is a flowchart of a method for playing a voice signal according to the present invention, and the method includes the following steps:

Step 701: Acquire a mixing signal of a maximum N-party site and a position information of a site with the largest energy in each frequency band at each time;

If the site number and maximum of the site with the largest energy in the largest N-party site are received in the above steps,

For the location information of the N-party site, the location information of the site with the largest energy in the largest N-party site is determined from the location information of the largest N-party site based on the site number.

In the above steps, when the received data is a mixed code stream and a position information stream, the mixed signal and the position information stream are decoded to obtain the mixed signal and each frequency band at each time. The orientation information of the site with the largest energy; when the received data is the mixed signal stream and the orientation information of the site with the largest energy in each frequency band at each time, by decoding the mixed code stream, the solution is obtained. The mixed signal is obtained, and finally the mixed signal and the orientation information of the site with the largest energy in each frequency band are obtained. Step 702: According to the correspondence between the auditory space parameter and the orientation information of the playback device, Obtaining an auditory spatial parameter of the playback device corresponding to the orientation information of the site with the highest energy on each frequency band at each time;

In the above steps, the auditory spatial parameters of the playback device include a level parameter and a delay parameter. The specific implementation process of step 902 may be: firstly setting a level parameter and a delay parameter corresponding to the azimuth information for the playback device, and acquiring, in step 701, the orientation information of the site with the highest energy in each frequency band at each time. After that, the corresponding relationship between the orientation information set by the playback device and the level parameter and the delay parameter is queried, and the level parameter of the playback device corresponding to the orientation information of the site with the largest energy at each time band is obtained. And delay parameters.

For example, in the conference site, two speakers are used as the playback device, and the position information of the site with the largest energy in a certain frequency band acquired is the upper left, and the level parameters and delay parameters of the two speakers can be obtained as follows: 1) Speaker 1 Level parameter in the upper left; 2) Level parameter in the upper left of speaker 2; 3) Delay parameter in the upper left of speaker 1; 4) Delay parameter in the upper left of speaker 2.

Step 703: Adjust the mixing signal by using the auditory space parameter of the playing device to play the adjusted mixing signal.

Wherein, the time-frequency conversion of the mixed signal is first performed, and the mixed signal in the time domain is converted into the mixed signal in the frequency domain, and the orientation information corresponding to the site with the largest energy in each frequency band is obtained. After playing the auditory spatial parameters of the device, the level and delay of the mixing signal in the frequency domain are adjusted by using the auditory spatial parameters of the playing device in each frequency band. Please refer to Figure 8 and Figure 8 for the adjustment of the auditory space parameters of the playback device in each frequency band. After adjusting the mixing signal on each frequency band, the adjusted mixing signal is inversely transformed in time and frequency, and the mixing signal in the frequency domain is converted into a mixing signal in the time domain, and finally in the time domain. The mix signal is played through the playback device. Embodiment 4 Corresponding to the above-mentioned method for playing a voice signal, the embodiment of the present invention further provides a device for playing a voice signal. Referring to FIG. 9, FIG. 9 is a structural diagram of a playback apparatus for a voice signal according to the present invention. The apparatus includes an acquisition unit 901, a spatial parameter obtaining unit 902, and an adjustment unit 903. The internal structure and connection relationship will be further described below in conjunction with the working principle of the device.

The obtaining unit 901 is configured to acquire a mixing signal of a maximum N-party venue and each frequency band at each moment The orientation information of the site with the largest energy;

The spatial parameter obtaining unit 902 is configured to obtain, according to the correspondence between the auditory spatial parameter and the orientation information of the playback device, the auditory space of the playback device corresponding to the orientation information of the site with the largest energy in each frequency band at each time. Parameter

The adjusting unit 903 is configured to adjust the mixed signal by using an auditory space parameter of the playing device, so as to play the adjusted mixed signal.

The obtaining unit 901 may include:

a first receiving unit 904, configured to receive a mixed code stream and a position information stream; and

The first decoding unit 905 is configured to decode the mixed code stream and the azimuth information code stream to obtain the mixed signal and the orientation information of the site with the largest energy in each frequency band at each time.

The first receiving unit 904 may be replaced by a second receiving unit, configured to receive the mixed code stream and the orientation information of the site with the largest energy in each frequency band at each time; the first decoding unit 905 may be replaced with the second decoding. And a unit, configured to decode the mixed code stream to obtain the mixed signal.

In this embodiment, the obtaining unit 901 may further include a first receiving unit, a first decoding unit, and a second receiving unit, and a second decoding unit.

The spatial parameter obtaining unit 902 may include:

The auditory spatial parameter preset unit 906 is configured to preset a level parameter and a delay parameter corresponding to the orientation information for the playback device; and

The query unit 907 is configured to query a correspondence between the orientation information and the level parameter and the delay parameter, and obtain a level parameter corresponding to the orientation information of the site with the largest energy in each frequency band at each moment. Delay parameter.

It should be noted that Fig. 9 does not define a complete structural diagram of a playback apparatus for a voice signal, but merely highlights the part involved in the inventive aspect of the present invention. It will be clear to those skilled in the art that the playback device of the voice signal should also include a player, and the adjusted mix signal output by the adjustment unit 903 is used as an input signal of the player. The player plays the adjusted mix signal.

It can be seen from the foregoing embodiment of the present invention that when the voice signal is processed, the orientation information is set in advance for all the sites participating in the conference, and in the largest N-party conference site, the orientation of the site with the largest energy in each frequency band is determined. Information that transmits the orientation information along with the mixing signal. In the voice letter When playing the number, according to the corresponding relationship between the received orientation information and the spatial parameters of the playback device, the spatial parameters of each playback device at the playback end are obtained, and the spatial parameters of the playback device are used to adjust the mixing signal, which will be adjusted. When the rear mixing signal is played, the auditory space of the sound source can be reconstructed in the venue, so that the sound of the largest N-party venue has a spatial stereoscopic feeling during playback, and the user can clearly understand the sound of each of the largest N-party venues, and further increases the sound. The user's experience of the spot experience.

In addition, when the site with the largest energy is in multiple pictures, the orientation information of the site with the largest energy will change accordingly with the change of the orientation of the site in the multi-picture, so that the orientation of the source is made when the voice signal is played. The orientation of the images is consistent, further increasing the user's experience of the presence experience.

The foregoing detailed description of the processing, playback method, and apparatus for a voice signal provided by the present invention is only for facilitating understanding of the method and core idea of the present invention. Meanwhile, for those of ordinary skill in the art, The present invention is not limited by the scope of the present invention.

Claims

Rights request

A method for processing a voice signal, the method comprising:

According to the orientation information set for the meeting place of the conference, in the largest N-party venue, the orientation information of the site with the largest energy in each frequency band at each moment in the mixed signal is determined;

The mixing signal of the largest N-party site and the orientation information of the site with the largest energy in each frequency band at each time are sent to the conference terminal participating in the conference.

The method according to claim 1, wherein the orientation information set for the meeting place participating in the meeting determines the orientation of the site with the largest energy in each frequency band at each time in the largest N-party meeting site. Information includes:

According to the order of joining, the orientation is preset for the venue where the conference is attended, and the preset orientation information is obtained;

Comparing the energy values of the speech signals of the largest N-party site at each time in each frequency band, and obtaining the site with the largest energy in each frequency band at each time;

When the site with the largest energy is not in the multi-screen, the orientation information of the site with the largest energy is set according to the preset orientation information. When the site with the largest energy is in the multi-screen, the site with the largest energy is set according to the multi-screen orientation information. Bearing information.

According to the order of joining, the orientation of the largest N-party venue is preset in order, and the preset orientation information of the largest N-party venue is obtained;

The method according to claim 1, wherein the transmitting the location information of the maximum N-party site and the location information of the site with the largest energy in each frequency band at each time to the conference site includes: And encoding the sound mixing signal and the orientation information of the site with the largest energy in each frequency band at each time to obtain a mixed code stream and a position information code stream respectively, and the mixed code stream and the position information stream Send to the venue terminal that attended the conference.

The method according to claim 4, wherein the manner of encoding the mixed signal of the largest N-party venue and the orientation information of the site with the largest energy in each frequency band at each time is different from the hybrid The way the signal is encoded.

The method according to claim 1, wherein the transmitting the location information of the maximum N-party venue and the location information of the site with the largest energy in each frequency band at each time to the conference site includes:

The mixed signal is encoded to obtain a mixed code stream, and the azimuth information of the mixed code stream and the site with the highest energy in each frequency band is transmitted to the conference terminal participating in the conference.

The method according to claim 1, wherein the transmitting the location information of the maximum N-party site and the location information of the site with the largest energy in each frequency band at each time to the conference site includes:

And encoding the mixed signal to obtain a mixed code stream, and sending the mixed code stream and the number of the site with the largest energy in each frequency band at each time and the position information of the largest N-party site to the conference. The venue terminal.

8. A method for playing a voice signal, the method comprising:

Obtaining the mixing signal of the largest N-party site and the orientation information of the site with the largest energy in each frequency band at each time;

Obtaining an auditory spatial parameter of the playback device corresponding to the orientation information of the site with the highest energy in each frequency band at each moment according to the correspondence between the auditory spatial parameter and the orientation information of the playback device;

The mix signal is adjusted using the auditory spatial parameters of the playback device, and the adjusted mix signal is played.

The method according to claim 8, wherein the acquiring the mixing signal of the largest N-party site and the location information of the site having the largest energy in each frequency band at each time include:

Receiving a mixed code stream and a position information stream;

Decoding the mixed code stream and the position information code stream to obtain the mixed signal and each moment The orientation information of the site with the highest energy in each band.

The method according to claim 8, wherein the acquiring the sounding signal and the orientation information of the site having the largest energy in each frequency band at each time comprises:

Receiving the mixed code stream and the orientation information of the site with the largest energy in each frequency band;

Decoding the mixed code stream to obtain the mixed signal.

The method according to claim 8, wherein the auditory spatial parameters of the playback device comprise: a level parameter and a delay parameter.

The method according to claim 11, wherein, according to the correspondence between the auditory spatial parameter and the orientation information of the playback device, the orientation of the site with the largest energy in each frequency band at each time is obtained. The auditory spatial parameters of the playback device corresponding to the information include: querying the correspondence between the orientation information preset by the playback device and the level parameter and the delay parameter, and obtaining the conference with the largest energy in each frequency band at each moment. The orientation information corresponds to the level parameter and the delay parameter.

The method according to claim 8, wherein the acquiring the mixed signal of the largest N-party site and the location information of the site having the largest energy in each frequency band at each time include:

Receiving the mixed code stream and the number of the site with the largest energy in each frequency band at each time and the orientation information of the largest N-party site;

Decoding the mixed code stream and the azimuth information code stream to obtain the mixed signal; obtaining each time every time according to the number of the site with the largest energy and the position information of the largest N-party site in each frequency band at each time The position information of the site with the highest energy in the frequency band.

14. A device for processing a voice signal, the device comprising:

The position determining unit is configured to determine, according to the orientation information set for the meeting place of the meeting, the orientation information of the site with the largest energy in each frequency band at each time in the mixed signal in the maximum N-party meeting place; The mixing signal of the largest N-party site and the orientation information of the site with the largest energy in each frequency band at each time are sent to the conference terminal participating in the conference.

The apparatus according to claim 14, wherein the orientation determining unit comprises: a first orientation presetting unit, configured to pre-set an orientation for the conference site participating in the conference according to an order of enrollment, and obtain a preset Bearing information a comparison unit, configured to compare energy values of each frequency band of the voice signal of the largest N-party site at each time, and obtain a site with the largest energy in each frequency band at each time;

a first setting unit, configured to: when the site with the largest energy is not in the multi-screen, set the orientation information of the site with the largest energy according to the preset location information;

The second setting unit is configured to set the orientation information of the site with the largest energy according to the multi-screen orientation information when the site with the largest energy is in the multi-screen.

The device according to claim 14, wherein the orientation determining unit comprises: a second orientation preset unit, configured to pre-set the orientation for the largest N-party venue according to the order of joining, to obtain a maximum N-party venue. Preset orientation information;

a comparison unit, configured to compare energy values of each frequency band of the voice signal of the largest N-party site at each time, and obtain a site with the largest energy in each frequency band at each time;

a third setting unit, configured to: when the site with the largest energy is not in the multi-screen, set the orientation information of the site with the largest energy according to the preset location information;

The fourth setting unit is configured to set the orientation information of the site with the largest energy according to the multi-screen orientation information when the site with the largest energy is in the multi-screen.

The device according to any one of claims 14-16, wherein the transmitting unit comprises:

a first sending unit, configured to encode the sound mixing signal and the orientation information of the site with the largest energy in each frequency band at each time, respectively, to obtain a mixed code stream and a position information code stream, respectively, to mix the sound The code stream and the direction information stream are sent to the conference terminal participating in the conference;

and / or,

a second sending unit, configured to: encode the mixed signal to obtain a mixed code stream, and send the mixed code stream and the position information of the site with the largest energy in each frequency band at each time to the conference The venue terminal.

18. A playback device for a voice signal, the device comprising:

An acquiring unit, configured to acquire a mixing signal of a maximum N-party site and a position information of a site with the largest energy amount in each frequency band at each time;

a spatial parameter obtaining unit, configured to obtain, according to a correspondence between the auditory spatial parameter and the azimuth information of the playback device, corresponding to the orientation information of the site with the largest energy in each frequency band at each time Auditory spatial parameters of the playback device;

And an adjusting unit, configured to adjust the mixing signal by using an auditory space parameter of the playing device, so as to play the adjusted mixing signal.

The device according to claim 18, wherein the acquiring unit comprises: a first receiving unit, configured to receive a mixed code stream and a position information code stream;

And a first decoding unit, configured to decode the mixed code stream and the azimuth information code stream, to obtain the mixed signal and the orientation information of the site with the largest energy in each frequency band at each time.

The device according to claim 18, wherein the acquiring unit comprises: a second receiving unit, configured to receive the mixed code stream and the orientation information of the site with the largest energy in each frequency band at each time ;

And a second decoding unit, configured to decode the mixed code stream to obtain the mixed signal.