WO2013167066A2

WO2013167066A2 - Method and device for regulating communication voice

Info

Publication number: WO2013167066A2
Application number: PCT/CN2013/079809
Authority: WO
Inventors: 吴佩莉
Original assignee: 中兴通讯股份有限公司
Priority date: 2013-03-08
Filing date: 2013-07-22
Publication date: 2013-11-14
Also published as: CN104038610A; WO2013167066A3

Abstract

A method and device for regulating communication voice, wherein the method comprises: accept a result signal of the voice quality detection (101), when the result signal indicates the current voice of poor quality, further judge that whether the current communicating contact person is the first contact person (102), if the judging result is not satisfied, search the mapping relation between the stored contact person and the audio feature parameters, and obtain the corresponding audio feature parameter to the contact person, then regulate the communication voice according to the audio feature parameter. Using the method and device for regulating communication voice can regulate communication voice real-timely, improve the quality of communication voice, and enhance the user experience.

Description

Call voice adjustment method and device

The present invention relates to a voice adjustment technique in wireless communication, and in particular to a call voice adjustment method and apparatus. Background technique

At present, the voice quality of the call of the mobile terminal is completed by the audio processing module. Generally, the audio processing module mainly includes: an analog gain module, a digital gain module, an automatic gain control module, a finite impulse response or an infinite impulse response filter, and modules such as echo suppression and noise cancellation.

The traditional solution for audio processing of existing mobile phones is: setting multiple usage modes, such as handheld mode, headset mode, hands-free mode or Bluetooth mode; and setting corresponding audio feature parameters in various modes to ensure call voice quality Meet the needs of users in different environments.

However, although multiple usage modes are considered, the audio feature parameters are fixed in the same usage mode, which allows different users to use the fixed audio feature parameters after different audio frequency characteristics or speech characteristics. The processing module weakens the differences or characteristics of the users, and cannot respond to the voice changes in real time in a timely manner. The processing capability is limited, so that the voice quality of the call cannot be completely guaranteed, which affects the user experience. Summary of the invention

In view of this, the main purpose of the embodiments of the present invention is to provide a method and device for adjusting a call voice, which can adjust call voices in real time, improve call voice quality, and improve user experience.

To achieve the above objective, the technical solution of the embodiment of the present invention is implemented as follows:

The embodiment of the invention provides a method for adjusting a call voice, and the method includes: Receiving a result signal of the voice quality detection;

When the result signal indicates that the current voice quality is poor, further determining whether the current call contact is the contact of the first call; when the result of the determination is no, retrieving the mapping relationship between the stored contact and the audio feature parameter, acquiring the location Describe the audio feature parameters corresponding to the contact;

The call voice is adjusted according to the audio feature parameters.

Preferably, the method further comprises: when the result signal indicates that the current voice quality is poor and the current call contact is the first-time contact, the call voice is adjusted according to the preset audio feature parameter.

Preferably, the method further includes: when the result signal indicates that the current voice quality is good and the current call contact is a first-time call, extracting audio feature parameters of the voice, and storing the contact and audio features The parameter, the mapping relationship between the contact and the detection parameter in the result signal.

Preferably, the method further includes: when the result signal indicates that the current voice quality is good and the current call contact is not the first call contact, retrieving the mapping relationship between the contact and the detection parameter in the result signal, and obtaining The detection parameter corresponding to the contact, if the voice quality represented by the current result signal is better than the voice quality represented by the stored detection parameter, extracting an audio feature parameter of the current voice, and updating an audio feature corresponding to the contact The parameter and the detection parameter corresponding to the contact.

Preferably, before the receiving the result signal of the voice quality detection, the method further includes: detecting the input voice data packet, sending the voice data packet to the voice quality detecting device on the network side for voice quality detection, and receiving the returned voice The resulting signal for voice quality detection.

The embodiment of the present invention further provides a call voice adjustment apparatus, where the apparatus includes: a transceiver module, a judgment module, a retrieval module, and a processing module;

The transceiver module is configured to receive a result signal of the voice quality detection, and when the result signal indicates that the current voice quality is poor, triggering the determining module; The determining module is configured to determine whether the current call contact is a contact for the first call, and when the determination result is no, trigger the retrieval module;

The search module is configured to retrieve a mapping relationship between the stored contact and the audio feature parameter, obtain an audio feature parameter corresponding to the contact, and send the audio feature parameter to the processing module;

The processing module is configured to adjust the call voice according to the audio feature parameter sent by the retrieval module.

Preferably, the processing module is further configured to adjust the call voice according to a preset audio feature parameter when the current call contact is a contact of the first call.

Preferably, the device further includes: an extraction module and a storage module; wherein

The extracting module is configured to: when the result signal indicates that the current voice quality is good, extract an audio feature parameter of the voice, and send the audio feature parameter and the corresponding contact to the storage module;

The storage module is configured to store a mapping relationship between the contact and audio feature parameters, the contact, and detection parameters in the result signal.

Preferably, the device further includes a comparison module configured to: when the current call contact is not the first call contact, obtain the current contact according to the mapping relationship between the contact retrieved by the retrieval module and the detection parameter in the result signal. The corresponding detection parameter of the person compares the voice quality represented by the current result signal with the voice quality indicated by the retrieved detection parameter, and sends the comparison result to the storage module;

Correspondingly, the storage module is further configured to: when the comparison result sent by the comparison module is that the voice quality indicated by the current result signal is good, update the stored audio feature parameter corresponding to the contact and the corresponding to the contact Detection parameters.

Preferably, the device further includes a detecting module configured to detect the input voice data packet, and send the voice data packet to the transceiver module; Correspondingly, the transceiver module is configured to send the voice data packet sent by the detection module to the voice quality detecting device on the network side.

The method and device for adjusting the call voice provided by the embodiment of the present invention receive the result signal of the voice quality detection. When the result signal indicates that the current voice quality is poor, it is further determined whether the current call contact is the contact of the first call. The contact of the current call is not the contact of the first call, and the mapping relationship between the stored contact and the audio feature parameter is retrieved, the audio feature parameter corresponding to the contact is obtained, and the call voice is adjusted according to the audio feature parameter. In this way, the voice of the call can be adjusted according to different voice frequency characteristics or voice characteristics of the user, which improves the voice quality of the call and improves the user experience. DRAWINGS

1 is a schematic flowchart of a call voice adjustment method according to a first embodiment of the present invention; FIG. 2 is a schematic structural diagram of a call voice adjustment apparatus according to a second embodiment of the present invention; A schematic diagram of the flow of the call voice adjustment method. detailed description

The embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. FIG. 1 is a schematic flowchart of a method for adjusting a call voice according to a first embodiment of the present invention. As shown in FIG. 1, the method includes the following steps:

Step 101: Receive a result signal of the voice quality detection. When the result signal indicates the current voice quality difference, perform step 102:

Here, before the receiving the result signal of the voice quality detection, the method further includes: detecting the input voice data packet, and transmitting the voice data packet to the voice quality detecting device on the network side for voice quality detection, and receiving the returned voice quality Result signal of detection;

The voice quality detection device on the network side performs voice quality detection according to the method for detecting voice quality by the existing network side device, and sets a standard value or a numerical range to detect the voice. Whether the sound quality is good, and the detection result is good or bad is signaled back to the terminal; since the specific voice quality detection method is prior art, it will not be described here;

Specific detection parameters may include: signal to noise ratio, distortion, noise transmission, and degree of suppression, etc. parameters capable of indicating voice quality;

The detection result includes parameters such as a signal to noise ratio, a distortion degree, a noise transmission, and a suppression degree in the current voice data;

The signal mode can be high or low, or different values.

Step 102: Determine whether the contact of the current call is the contact of the first call. If the result of the determination is no, perform step 103; if the result of the determination is yes, adjust the call voice according to the preset audio feature parameter.

Step 103: Retrieving a mapping relationship between the stored contact and the audio feature parameter, acquiring an audio feature parameter corresponding to the contact, and adjusting the call voice according to the audio feature parameter; where the audio feature parameter includes a spectrum combination, a parameter such as a spectral envelope that can characterize a speech feature;

Specifically, the finite impulse response or the wireless impulse response filter is mainly set according to the spectrum combination in the audio feature parameters, and the low frequency component is raised when the low frequency component is insufficient, and the high frequency component is added at the corresponding frequency point when the high frequency component is missing. The high frequency noise portion is suppressed such that the spectral characteristics in the response curve and the audio characteristic parameters are as close as possible.

Preferably, the method further includes: when the result signal indicates that the current voice quality is poor and the current call contact is the contact of the first call, adjusting the call voice according to the preset audio feature parameter;

Here, the preset audio feature parameter is a default audio feature parameter preset in the usage mode in a certain usage mode, such as a handheld mode, a headset mode, a hands-free mode, and the like.

Preferably, the method further comprises: when the result signal indicates that the current voice quality is good and When the contact of the previous call is the contact of the first call, the audio feature parameter of the voice is extracted, and the mapping relationship between the contact and the audio feature parameter, the contact parameter in the contact and the result signal is stored.

Preferably, the method further includes: when the result signal indicates that the current voice quality is good and the current call contact is not the first call contact, retrieving the mapping relationship between the contact and the detection parameter in the result signal, and obtaining The detection parameter corresponding to the contact, if the voice quality represented by the current result signal is better than the voice quality indicated by the stored detection parameter, extracting the audio feature parameter of the voice, and updating the audio corresponding to the contact Feature parameters and detection parameters corresponding to the contact, to optimize audio feature parameters, improve voice fidelity, and control data storage capacity;

Specifically, the parameters such as signal-to-noise ratio, distortion, noise transmission, and suppression degree of the detection parameters may be weighted by a predetermined degree of importance, and two average values are counted, and the voice quality of the call is compared according to the average value. The comparison process is prior art, and details are not described herein again.

Preferably, the method further includes: periodically clearing a mapping relationship between the stored contact and the audio feature parameter, and clearing data of the contact with few contacts; wherein the contact with few contacts may be at a set time Within the range, set the number of contacts with less than N calls.

2 is a schematic structural diagram of a call voice adjustment apparatus according to a first embodiment of the present invention. As shown in FIG. 2, the apparatus includes a transceiver module 21, a determination module 22, a retrieval module 23, and a processing module 24;

The transceiver module 21 is configured to receive a result signal of the voice quality detection, and when the result signal indicates that the current voice quality is poor, the triggering module 22 is triggered;

The determining module 22 is configured to determine whether the contact of the current call is the contact of the first call, and when the determination result is no, the retrieval module 23 is triggered;

The searching module 23 is configured to retrieve a mapping relationship between the stored contact and the audio feature parameter, obtain an audio feature parameter corresponding to the contact, and send the audio feature parameter to the location Management module 24;

The processing module 24 is configured to adjust the call voice according to the audio feature parameters sent by the retrieval module 23.

Preferably, the processing module 24 is further configured to adjust the call voice according to a preset audio feature parameter when the current call contact is the first call contact.

Preferably, the device further includes: an extracting module 26 and a storage module 25; wherein the extracting module 26 is configured to: when the result signal indicates that the current voice quality is good, extract audio feature parameters of the voice, and The audio feature parameters and corresponding contacts are sent to the storage module;

The storage module 25 is configured to store a mapping relationship between the contact and the audio feature parameter, the contact, and the detection parameter in the result signal.

Preferably, the device further includes a comparison module 27 configured to obtain, according to a mapping relationship between the contact retrieved by the retrieval module 23 and the detection parameter in the result signal, when the current call's contact is not the first-time contact. The detection parameter corresponding to the current contact, the voice quality represented by the current result signal is compared with the voice quality indicated by the retrieved detection parameter, and the comparison result is sent to the storage module 25;

The storage module 25 is further configured to: when the comparison result sent by the comparison module is that the voice quality is good, update the stored audio feature parameter corresponding to the contact and the detection parameter corresponding to the contact.

Preferably, the device further includes a detecting module 28 configured to detect the input voice data packet, and send the voice data packet to the transceiver module 21;

The transceiver module 21 is configured to send the voice data packet sent by the detecting module 28 to the voice quality detecting device on the network side.

FIG. 3 is a schematic flowchart of a method for adjusting a call voice according to a second embodiment of the present invention. As shown in FIG. 3, the method includes the following steps: Step 301 to step 303: detecting the input voice data packet, transmitting the voice data packet to the voice quality detecting device on the network side for voice quality detection, receiving the returned voice quality detection result signal, and when the result signal indicates the current voice quality If yes, step 304 to step 309 are performed; when the result signal indicates that the current voice quality is poor, steps 310 to 313 are performed.

Here, the voice quality detection device of the network side performs voice quality detection according to a method for detecting voice quality by an existing network side device, and sets a standard value or a value range to detect whether the voice quality is good.

Among them, the specific detection parameters include signal-to-noise ratio, distortion, noise transmission, and degree of suppression, etc., which can indicate the voice quality.

Step 304: Determine whether the contact of the current call is the contact of the first call. If the result of the determination is yes, perform steps 305 to 306; if the result of the determination is no, perform steps 307 to 309.

Step 305 to step 306: Extract the audio feature parameters of the voice, store the mapping relationship between the contact and the audio feature parameters, and end the processing flow.

Step 307 to step 308: Retrieving a mapping relationship between the contact and the detection parameter in the result signal, obtaining a detection parameter corresponding to the current contact, and determining whether the voice quality represented by the current result signal is greater than the voice represented by the stored detection parameter. The quality is good. If the result of the judgment is yes, go to step 309, otherwise no operation will be performed.

Step 309: Extract the audio feature parameters of the voice, update the stored audio feature parameters corresponding to the contact, and the detection parameters corresponding to the contact, and end the process.

Here, if the current call contact is not the first call contact, the stored detection parameter corresponding to the current contact is compared with the detection parameter in the current result signal, and the signal to noise ratio, distortion, and noise in the detected parameter may be specifically detected. The parameters such as the transmission and the suppression degree are weighted by a predetermined degree, and the average value is calculated. The voice quality of the call is compared according to the average value. The comparison process is prior art and will not be described here. If the current call voice quality If the quantity is better, the stored audio feature parameters corresponding to the contact and the detection parameters corresponding to the contact are updated to optimize the audio feature parameters, improve the fidelity of the voice, and control the data storage capacity.

The audio feature parameters include a spectrum combination, a spectrum envelope, and the like that can characterize a speech feature.

Step 310: Determine whether the contact of the current call is the contact of the first call. If the result of the determination is yes, go to step 311; if the result of the determination is no, go to step 312 to step 313.

Step 311: Adjust the call voice according to the preset audio feature parameters, and end the process.

Step 312 to step 313: Retrieve the mapping relationship between the stored contact and the audio feature parameter, obtain the audio feature parameter corresponding to the contact, and adjust the call voice according to the audio feature parameter.

The above is only the preferred embodiment of the present invention and is not intended to limit the scope of the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and scope of the invention are intended to be included within the scope of the invention. Industrial applicability

The mapping relationship between the stored contact and the audio feature parameter is performed by the embodiment of the present invention. When the result signal of the received voice quality detection indicates that the current voice quality is poor, the stored contact is retrieved. The mapping relationship between the audio feature parameters is obtained, the audio feature parameters corresponding to the contact are obtained, and the call voice is adjusted according to the audio feature parameter. In this way, the voice of the call can be adjusted according to different voice frequency characteristics or voice characteristics of the user, which improves the voice quality of the call and improves the user experience.

Claims

claims

1. A call voice adjustment method, the method includes:

Receive the result signal of voice quality detection;

When the result signal indicates that the current voice quality is poor, it is further judged whether the contact of the current call is the contact of the first call; when the result of the judgment is no, the mapping relationship between the stored contact and the audio feature parameters is retrieved, and all the The audio feature parameters corresponding to the contact;

The call voice is adjusted according to the audio characteristic parameters.

2. The call voice adjustment method according to claim 1, wherein the method further includes: when the result signal indicates that the current voice quality is poor and the contact of the current call is the contact of the first call, press the preset The audio feature parameters adjust the call voice.

3. The call voice adjustment method according to claim 1, wherein the method further includes: when the result signal indicates that the current voice quality is good and the contact of the current call is the first call, extracting the voice audio feature parameters, and stores the mapping relationship between the contact and the audio feature parameter, and the contact and the detection parameter in the result signal.

4. The call voice adjustment method according to claim 3, wherein the method further includes: when the result signal indicates that the current voice quality is good and the contact of the current call is not the contact of the first call, retrieving the contact The mapping relationship between the person and the detection parameters in the result signal is to obtain the detection parameters corresponding to the contact. If the voice quality represented by the current result signal is better than the voice quality represented by the stored detection parameters, extract the current voice Audio feature parameters: update the audio feature parameters corresponding to the contact and the detection parameters corresponding to the contact.

5. The call voice adjustment method according to claim 1, wherein before receiving the result signal of the voice quality detection, the method further includes:

Detect the input voice data packet, send the voice data packet to the voice quality detection device on the network side for voice quality detection, and receive the returned voice quality detection result signal.

6. A call voice adjustment device, the device includes: a transceiver module, a judgment module, a detection module search module and processing module; among them,

The transceiver module is configured to receive the result signal of the voice quality detection, and when the result signal indicates that the current voice quality is poor, the judgment module is triggered;

The judgment module is configured to judge whether the contact of the current call is the contact of the first call, and when the judgment result is no, trigger the retrieval module;

The retrieval module is configured to retrieve the mapping relationship between stored contacts and audio feature parameters, obtain the audio feature parameters corresponding to the contact, and send the audio feature parameters to the processing module;

The processing module is configured to adjust the call voice according to the audio feature parameters sent by the retrieval module.

7. The call voice adjustment device according to claim 6, wherein the processing module is further configured to adjust the call voice according to the preset audio feature parameters when the current contact is the first call contact. .

8. The call voice adjustment device according to claim 6, wherein the device further includes: an extraction module and a storage module; wherein,

The extraction module is configured to extract the audio feature parameters of the voice when the result signal indicates that the current voice quality is good, and send the audio feature parameters and the corresponding contact to the storage module;

The storage module is configured to store the mapping relationship between the contact and the audio feature parameter, and the contact and the detection parameter in the result signal.

9. The call voice adjustment device according to claim 8, wherein the device further includes a comparison module configured to, when the contact of the current call is not the contact of the first call, the contact retrieved by the retrieval module and the The detection parameters corresponding to the current contact are obtained from the mapping relationship of the detection parameters in the result signal, compare the voice quality represented by the current result signal with the voice quality represented by the retrieved detection parameter, and send the comparison result to the storage module; The storage module is further configured to update the stored audio feature parameters corresponding to the contact and the detection parameters corresponding to the contact when the comparison result sent by the comparison module is that the voice quality represented by the current result signal is good.

10. The call voice adjustment device according to claim 6, wherein the device further includes a detection module configured to detect the input voice data packet and send the voice data packet to the transceiver module;

The transceiver module is configured to send the voice data packet sent by the detection module to the voice quality detection device on the network side.