A voice quality evaluation method, device and readable storage medium

Abstract

The application provides a voice quality evaluation method, device and readable storage medium, the method comprises the following steps: obtaining SIP signaling corresponding to a target call record and RTP packet data corresponding to the SIP signaling; identifying the first packet and the last packet of the RTP packet according to the SIP signaling and the RTP packet data; identifying the first packet, the last packet and all RTP packets between the first packet and the last packet in the RTP packet data as valid RTP packets; evaluating the voice quality of the target call record according to the time interval between the valid RTP packets and the time interval between the last packet and the voice call end identifier, and obtaining the evaluation result of the voice quality. The method, device and medium can solve the problem that the existing voice quality evaluation method only evaluates the voice quality according to the relationship between the RTP packets, ignores the validity of the RTP packets and the process judgment of the call start and end stage, and thus leads to the inaccurate evaluation result.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of communication, in particular to a voice quality evaluation method, device and readable storage medium. BACKGROUND

[0002] With the popularization of 5G network, people have higher expectations for more stable call quality, so the optimization of voice quality evaluation technology is particularly important.

[0003] In the research and practice process of the prior art, the inventor found that: the existing voice quality evaluation method only considers the relationship between RTP (Real-time Transport Protocol, Real-time Transport Protocol) packets as the basis for judging whether there is a voice problem, without judging the effectiveness of the RTP packet and the call end process. For the voice call start stage, the existing network has a terminal supporting the early media (P-Early-Media) function. When the caller starts calling and receives the 180Ring message, the caller starts sending uplink RTP messages, which can include audio or video content, such as public service advertisement animations in video ringback tone services, to maintain the connection active state or transmit media content in advance. These early media messages are sent before the called party answers the phone, and even if the called party does not answer the phone, the caller will continue to send a certain number of RTP packets as part of the keep-alive mechanism, resulting in inaccurate voice call quality evaluation. For the voice call end stage, the 200OK (BYES) message is generally received as a symbol of the voice call end process, and the judgment of no RTP packet transmission as the end of the call cannot accurately reflect the true end time of the voice call, resulting in inaccurate voice call quality evaluation. SUMMARY

[0004] The technical problem to be solved by the present application is to solve the above-mentioned shortcomings of the prior art, and to provide a voice quality evaluation method, device and readable storage medium, so as to solve the problem that the existing voice quality evaluation method only evaluates the voice quality according to the relationship between RTP packets, but ignores the effectiveness of the RTP packet and the process judgment of the call start and end stage, resulting in inaccurate evaluation results.

[0005] In a first aspect, the present application provides a voice quality evaluation method, which comprises:

[0006] Obtaining the session initiation protocol (SIP) signaling corresponding to the target call record and the real-time transport protocol (RTP) packet data corresponding to the SIP signaling;

[0007] Identifying the first packet and the last packet of the RTP packet according to the SIP signaling and the RTP packet data;

[0008] identify the first packet, the last packet and all the RTP packets between the first packet and the last packet in the RTP packet data as valid RTP packets;

[0009] evaluate the voice quality of the target call record according to the time interval between the valid RTP packets and the time interval between the last packet and the voice call end identifier, to obtain an evaluation result of the voice quality.

[0010] Further, the SIP signaling corresponding to the target call record and the RTP packet data corresponding to the SIP signaling are obtained, specifically including:

[0011] collect the SIP signaling and the RTP packet data generated by the user during the voice call;

[0012] match the collected SIP signaling and RTP packet data according to the timestamp information and the payload type, to obtain the correspondence between the SIP signaling and the RTP packet;

[0013] obtain the SIP signaling corresponding to the target call record and the RTP packet data corresponding to the SIP signaling according to the correspondence between the SIP signaling and the RTP packet.

[0014] Further, the collected SIP signaling includes a timestamp and SDP information, and the collected RTP packet data includes a sequence number, a timestamp, an SSRC and a payload type.

[0015] Further, the first packet and the last packet of the RTP packet are identified according to the SIP signaling and the RTP packet data, specifically including:

[0016] determine the time point of a first message in the SIP signaling as an indicator of a voice on time point as the voice on time point;

[0017] identify the first RTP packet after the voice on time point in the RTP packet data as the first packet;

[0018] determine the time point of a second message in the SIP signaling as an indicator of a voice end identifier as the voice end time point;

[0019] identify the last RTP packet before the voice end time point in the RTP packet data as the last packet.

[0020] Further, the voice quality of the target call record is evaluated according to the time interval between the valid RTP packets and the time interval between the last packet and the voice call end identifier, to obtain an evaluation result of the voice quality, specifically including:

[0021] calculate a first time interval between any two continuous RTP packets in the effective RTP packets;

[0022] calculate a second time interval between the tail packet and the voice call end identifier;

[0023] evaluate the voice quality of the target call record according to the first time interval and the second time interval, and obtain an evaluation result of the voice quality.

[0024] Further, the evaluation of the voice quality of the target call record according to the first time interval and the second time interval, and the obtaining of the evaluation result of the voice quality specifically include:

[0025] If any one of the first time intervals is greater than a preset first threshold or the second time interval is greater than a preset second threshold, it is determined that the evaluation result of the voice quality is that there is voice quality difference.

[0026] Further, the first threshold is 500 ms, and the second threshold is 1000 ms.

[0027] In a second aspect, the present application provides a voice quality evaluation device, which comprises:

[0028] a record data acquisition module, configured to acquire session initiation protocol (SIP) signaling corresponding to a target call record and real-time transport protocol (RTP) packet data corresponding to the SIP signaling;

[0029] a head-tail packet identification module, connected with the record data acquisition module, configured to identify a head packet and a tail packet of the RTP packet according to the SIP signaling and the RTP packet data;

[0030] an effective RTP packet identification module, connected with the head-tail packet identification module, configured to identify the head packet, the tail packet and all RTP packets between the head packet and the tail packet in the RTP packet data as effective RTP packets;

[0031] a voice quality evaluation module, connected with the effective RTP packet identification module, configured to evaluate the voice quality of the target call record according to time intervals between the effective RTP packets and a time interval between the tail packet and a voice call end identifier, and obtain an evaluation result of the voice quality.

[0032] In a third aspect, the present application provides a voice quality evaluation device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to realize the voice quality evaluation method of the first aspect.

[0033] In a fourth aspect, the present application provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the voice quality evaluation method in the first aspect.

[0034] The present application provides a voice quality evaluation method, device and readable storage medium. Firstly, the SIP signaling corresponding to the target call record and the RTP packet data corresponding to the SIP signaling are acquired; then, the first packet and the last packet of the RTP packet are identified according to the SIP signaling and the RTP packet data; then, the first packet, the last packet and all the RTP packets between the first packet and the last packet in the RTP packet data are identified as valid RTP packets; finally, the voice quality of the target call record is evaluated according to the time interval between the valid RTP packets and the time interval between the last packet and the voice call end identifier, and the evaluation result of the voice quality is obtained. The present application ensures that the voice quality evaluation is based on only the valid RTP packet data by preferentially judging the validity of the RTP packet. Meanwhile, the time interval of the valid RTP packet and the time interval between the last packet and the voice call end identifier are comprehensively analyzed, which significantly improves the accuracy and reliability of the voice quality evaluation, and makes the evaluation result closer to the actual call experience. The present application solves the problem that the existing voice quality evaluation method only evaluates the voice quality according to the relationship between the RTP packets, and ignores the validity of the RTP packet and the process judgment of the call start and end stage, thereby resulting in inaccurate evaluation result. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 The flow chart of a voice quality evaluation method of the present application embodiment 1;

[0036] Figure 2 The flow chart of the RTP first packet identification of the present application embodiment;

[0037] Figure 3 The flow chart of the RTP last packet identification of the present application embodiment;

[0038] Figure 4 The time point diagram of the RTP first packet and the 200OK (BYE) message of the present application embodiment;

[0039] Figure 5 The flow chart of another voice quality evaluation method of the present application embodiment;

[0040] Figure 6 The structural schematic diagram of a voice quality evaluation device of the present application embodiment 2;

[0041] Figure 7 The structural schematic diagram of a voice quality evaluation device of the present application embodiment 3. DETAILED DESCRIPTION

[0042] In order to better understand the technical solutions of the present application, the embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

[0043] It can be understood that the specific embodiments and drawings described herein are merely intended to explain the present application, but not to limit the present application.

[0044] It can be understood that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0045] It can be understood that, for the convenience of description, only the parts related to the present application are shown in the drawings of the present application, and the parts unrelated to the present application are not shown in the drawings.

[0046] It can be understood that each unit and module involved in the embodiments of the present application can correspond to only one entity structure, or can be composed of multiple entity structures, or multiple units and modules can be integrated into one entity structure.

[0047] It can be understood that the terms "first", "second" and the like in the embodiments of the present application are used to distinguish different objects, or to distinguish different treatments of the same object, but not to describe a specific order of the objects.

[0048] It can be understood that the functions and steps marked in the flowcharts and block diagrams of the present application can occur in an order different from that marked in the drawings without conflict.

[0049] It can be understood that in the flowcharts and block diagrams of the present application, the architecture, functions and operations of the possible implementations of the system, device, equipment and method according to the embodiments of the present application are shown. Each block in the flowchart or block diagram can represent a unit, module, program segment, code, which contains executable instructions for realizing the specified functions. Moreover, each block or combination of blocks in the block diagram and flowchart can be realized by a hardware-based system for realizing the specified functions, or by a combination of hardware and computer instructions.

[0050] It can be understood that the units and modules involved in the embodiments of the present application can be realized in the form of software or hardware, for example, the units and modules can be located in a processor.

[0051] Embodiment 1:

[0052] The embodiment provides a voice quality evaluation method, as shown in the figure, the method comprises the steps of: Figure 1

[0053] ​Step S101: Obtain SIP (Session Initiation Protocol) signaling corresponding to a target call record and RTP (Real-time Transport Protocol) packet data corresponding to the SIP signaling.

[0054] In this embodiment, the SIP signaling corresponding to the target call record and the RTP packet data corresponding to the SIP signaling can be obtained according to the SIP signaling and the RTP packet data collected and stored in advance.

[0055] Optionally, the SIP signaling corresponding to the target call record and the RTP packet data corresponding to the SIP signaling are obtained, and specifically include the following steps.

[0056] Collecting SIP signaling and RTP packet data generated by a user during voice communication;

[0057] Obtaining a correspondence between the SIP signaling and the RTP packet by matching the collected SIP signaling and RTP packet data according to timestamp information and payload type;

[0058] Obtaining the SIP signaling corresponding to the target call record and the RTP packet data corresponding to the SIP signaling according to the correspondence between the SIP signaling and the RTP packet.

[0059] In this embodiment, the SIP signaling and the RTP packet data generated by the user during voice communication can be collected by a core network, wherein the collected SIP signaling includes timestamp and SDP (Session Description Protocol) information, and the collected RTP packet data includes sequence number, timestamp, SSRC (Synchronization Source) and payload type.

[0060] In this embodiment, the correspondence between the SIP signaling and the RTP packet is obtained by matching and associating the timestamp and the SDP information carried in the SIP signaling with the timestamp and the payload type value carried in the RTP packet data according to the payload type and other information in the SDP information, and the SIP signaling and the RTP packet information of each call record are obtained, wherein the target call refers to any one of the call records.

[0061] Step S102: Identify the first packet and the last packet of the RTP packet according to the SIP signaling and the RTP packet data.

[0062] In the embodiment, the first packet and the last packet of the RTP packet are identified according to the SIP signaling and the RTP packet data.

[0063] Optionally, the first packet and the last packet of the RTP packet are identified according to the SIP signaling and the RTP packet data, and specifically include:

[0064] The time point of the first message in the SIP signaling indicating the speech connection is determined as the time point of the speech connection.

[0065] The first RTP packet after the time point of the speech connection in the RTP packet data is identified as the first packet.

[0066] The time point of the second message in the SIP signaling indicating the end of the speech call is determined as the time point of the speech end.

[0067] The last RTP packet before the time point of the speech end in the RTP packet data is identified as the last packet.

[0068] In the embodiment, taking the calling side as an example, the ACK message sent by the calling side SBC after receiving the 200 OK (invite) message sent by the calling terminal after the called terminal is off hook is taken as the first message indicating the speech connection, and the time point is the time point of the speech connection. The first RTP packet sent by the calling terminal after the calling side SBC receives the speech connection is taken as the first packet.

[0069] In the embodiment, taking the calling side as an example, the first 200 OK (BYE) message received by the calling side SBC is taken as the second message indicating the end of the speech call, and the time point is the time point of the speech end. The last RTP packet sent by the calling terminal after the calling side SBC receives the speech call is taken as the last packet.

[0070] Step S103: The first packet, the last packet and all the RTP packets between the first packet and the last packet in the RTP packet data are identified as the valid RTP packets.

[0071] Step S104: The speech quality of the target call record is evaluated according to the time interval between the valid RTP packets and the time interval between the last packet and the speech call end identifier, and the evaluation result of the speech quality is obtained.

[0072] In the embodiment, in order to improve the accuracy and reliability of the voice quality evaluation, the voice quality is evaluated by identifying the effective RTP packet data (including the first packet, the last packet and all RTP packets between the first packet and the last packet), combining the time interval between the effective RTP packets and the time interval between the last packet and the voice call end identifier. The method makes the evaluation result closer to the actual call experience.

[0073] Optionally, the voice quality of the target call record is evaluated according to the time interval between the effective RTP packets and the time interval between the last packet and the voice call end identifier, and an evaluation result of the voice quality is obtained, and specifically includes:

[0074] The first time interval between any two continuous RTP packets in the effective RTP packets is calculated.

[0075] The second time interval between the last packet and the voice call end identifier is calculated.

[0076] The voice quality of the target call record is evaluated according to the first time interval and the second time interval, and an evaluation result of the voice quality is obtained.

[0077] In the embodiment, the time when the calling side SBC receives the nth RTP packet is T(n), the time of the RTP last packet is T(m), and the time of the first 200OK (BYES) message is T(m+1). The time interval TR between the RTP and the time interval TB between the last packet and the first 200OK (BYES) message are:

[0078] TR=T(n)-T(n-1)

[0079] TB=T(m+1)-T(m)

[0080] The time interval between the RTP packets and the voice call end signaling is evaluated as the judgment basis of the voice quality evaluation.

[0081] Optionally, the voice quality of the target call record is evaluated according to the first time interval and the second time interval, and an evaluation result of the voice quality is obtained, and specifically includes:

[0082] If any one of the first time interval is greater than a preset first threshold or the second time interval is greater than a preset second threshold, it is judged that the evaluation result of the voice quality is that there is voice quality difference.

[0083] In the embodiment, since the process time is generally 500 ms in the process of resource release and the like involved in the call end stage, the time interval between the tail packet and the voice call end identifier is increased by 500 ms compared with the time interval between the RTP packets, therefore, the first threshold is preferably 500 ms, and the second threshold is preferably 1000 ms. That is, if TR is greater than 500 ms or TB is greater than 1000 ms, it is considered that there is voice quality difference.

[0084] Through statistical data analysis, the voice head and tail packet and signaling determination rules are added, and the single voice discontinuous statistical accuracy rate is increased from 92.66% to 96.35%, which is increased by 3.69pp. Compared with the voice quality evaluation accuracy rate without adding the head and tail packet and signaling determination rules, the voice quality evaluation accuracy rate is obviously improved.

[0085] It should be noted that the voice quality evaluation method provided by the embodiment of the application, by combining the voice call signaling process, constructing a comprehensive evaluation model of the voice call SIP signaling and the RTP packet, and preferentially judging the effectiveness of the RTP packet, the voice quality evaluation is performed according to the effective RTP packet data combined with the voice SIP signaling, and the accuracy of the voice perception evaluation judgment is improved.

[0086] In a specific embodiment, the voice quality evaluation method is applied to a voice quality evaluation system, the system is used for voice discontinuity, single voice and the like, and the voice perception evaluation result is output by correlatively evaluating and analyzing the RTP packet and the voice call signaling, the system specifically includes a data acquisition module, a data storage module, a call record correlation analysis module, an RTP head packet identification module, an RTP tail packet identification module and an evaluation module, and the modules are described as follows:

[0087] (1) The data acquisition module acquires the SIP signaling and the RTP packet data corresponding to the voice generated by the user in the voice call through the core network, wherein the SIP signaling includes but is not limited to the timestamp, the SDP information and the like, and the RTP packet data includes but is not limited to the packet sequence number, the timestamp, the SSRC (synchronization source), the Payload Type (payload type) and the like. The SDP is used for correlating the SIP signaling and the RTP packet data, and the SSRC is used for uniquely identifying the user call.

[0088] (2) The data storage module is used for storing the collected SIP signaling and RTP information related data.

[0089] (3) Call record association analysis module, the module carries the time stamp and SDP information in the SIP signaling, through the Payload Type (payload type) and other information in the SDP information, and the time stamp and PayloadType value carried in the RTP packet data are matched and associated, the corresponding relationship between the SIP signaling and the RTP packet is obtained, and the SIP signaling and the RTP packet information of each call record are obtained.

[0090] (4) RTP first packet identification module, the SIP signaling and the RTP packet information of the call record obtained are further confirmed to the first packet of the RTP packet. Figure 2 The flow chart of RTP first packet identification is shown, the ACK message sent by the called terminal after the 200OK (invite) message sent by the calling terminal is received by the calling side SBC as the voice connection identifier, the time point is the voice connection time point, and the first RTP packet sent by the calling terminal after the voice connection is received by the calling side SBC as the first packet.

[0091] (5) RTP tail packet identification module, the SIP signaling and the RTP packet information of the call record obtained are further confirmed to the tail packet of the RTP packet. Figure 3 The flow chart of RTP tail packet identification is shown, the first 200OK (BYE) message received by the calling side SBC is taken as the voice call end identifier, the time point is the voice end time point, and the last RTP packet sent by the calling terminal after the voice call is ended is received by the calling side SBC as the tail packet.

[0092] (6) Evaluation module, according to the voice SIP signaling and the RTP packet first and tail packet information obtained, the voice quality is evaluated through certain rules, and the evaluation result is output.

[0093] Taking the calling side as an example, Figure 4 The time point schematic diagram of the RTP first and tail packets and the 200OK (BYE) message is shown, the time when the nth RTP packet is received by the calling side SBC is T (n), the time of the tail packet is T (m), the time of the first 200OK (BYE) message is T (m+1), and the time interval TR between the RTP and the time interval TB between the tail packet and the first 200OK (BYE) message are:

[0094] TR=T (n)-T (n-1)

[0095] TB=T (m+1)-T (m)

[0096] By evaluating the time interval between the RTP packets and the signaling between the end of the voice call as the basis for voice quality evaluation, since the process time is generally 500ms in the end of the call stage involving resource release, etc., the tail packet and the first 200OK (BYE) message time interval is increased by 500ms compared to the time interval between the RTP packets, and if TR is greater than 500ms or TB is greater than 1000ms, it is considered that there is poor voice quality.

[0097] Through statistical data analysis, the voice first and last packet and signaling determination rule are added, and the accuracy rate of voice single on-off intermittent statistics is increased from 92.66% to 96.35%, which is increased by 3.69pp. Compared with the voice quality evaluation accuracy rate without adding the first and last packet and signaling determination rule, the voice quality evaluation accuracy rate is obviously improved.

[0098] It should be noted that in order to solve the problem of inaccurate evaluation in the existing voice quality evaluation method, the present application provides a voice evaluation scheme based on RTP packet and voice call signaling. The scheme considers the signaling process of voice on and off at the same time, judges the voice quality through SIP signaling and RTP packet, so as to realize more accurate and comprehensive evaluation.

[0099] Based on the above voice quality evaluation system, as shown in Figure 5 The corresponding voice quality evaluation method includes the following steps:

[0100] 1. The data acquisition module acquires voice call SIP signaling and RTP packet data every day;

[0101] 2. Save the collected SIP signaling and RTP packet data to the data storage server;

[0102] 3. Complete the association of SIP signaling and RTP packet data to obtain different call slips;

[0103] 4. Confirm the first packet of the RTP packet according to the slip data after association (i.e. confirm the first packet of the RTP according to the slip SIP signaling and RTP packet data time information);

[0104] 5. Confirm the tail packet of the RTP packet according to the slip data after association (i.e. confirm the tail packet of the RTP according to the slip SIP signaling and RTP packet data time information);

[0105] 6. Evaluate the time interval between the RTP packets and the time interval between the tail packet and the voice call end identifier, and output the evaluation result.

[0106] It should be noted that the voice quality evaluation method provided by the embodiment of the present application combines the voice RTP packet with the voice call signaling process, and more accurately judges the abnormal phenomena such as discontinuity of the first packet and the tail packet of the voice. Compared with the time delay comparison between the traditional RTP packets, it is more in line with the user's perception of abnormal phenomena such as call discontinuity.

[0107] The voice quality evaluation method provided by the embodiment of the present application first acquires the session initiation protocol (SIP) signaling corresponding to the target call record and the real-time transport protocol (RTP) packet data corresponding to the SIP signaling; then identifies the first packet and the tail packet of the RTP packet according to the SIP signaling and the RTP packet data; then identifies the first packet, the tail packet and all RTP packets located between the first packet and the tail packet in the RTP packet data as valid RTP packets; and finally evaluates the voice quality of the target call record according to the time interval between the valid RTP packets and the time interval between the tail packet and the voice call end identifier, to obtain the evaluation result of the voice quality. The present application ensures that the voice quality evaluation is based only on valid RTP packet data by prioritizing the validity judgment of the RTP packet. At the same time, the time interval of the valid RTP packet and the time interval between the tail packet and the voice call end identifier are combined for comprehensive analysis, which significantly improves the accuracy and reliability of the voice quality evaluation, making the evaluation result closer to the actual call experience. The existing voice quality evaluation method only evaluates the voice quality according to the relationship between the RTP packets, but ignores the validity of the RTP packets and the process judgment of the call start and end stage, resulting in inaccurate evaluation results.

[0108] Embodiment 2:

[0109] As shown in Figure 6 The present embodiment provides a voice quality evaluation device for executing the voice quality evaluation method described above, comprising:

[0110] The call record data acquisition module 11 is configured to acquire the session initiation protocol (SIP) signaling corresponding to the target call record and the real-time transport protocol (RTP) packet data corresponding to the SIP signaling;

[0111] The first and last packet identification module 12 is connected with the call record data acquisition module 11 and is configured to identify the first packet and the tail packet of the RTP packet according to the SIP signaling and the RTP packet data;

[0112] The valid RTP packet identification module 13 is connected with the first and last packet identification module 12 and is configured to identify the first packet, the tail packet and all RTP packets located between the first packet and the tail packet in the RTP packet data as valid RTP packets;

[0113] The voice quality evaluation module 14, connected with the valid RTP packet identification module 13, is configured to evaluate the voice quality of the target call record according to the time interval between the valid RTP packets and the time interval between the tail packet and the voice call end identifier, and obtain an evaluation result of the voice quality.

[0114] Optionally, the call record data acquisition module 11 comprises:

[0115] a data acquisition unit configured to acquire SIP signaling and RTP packet data generated by a user during a voice call;

[0116] a data matching unit configured to obtain a correspondence between the SIP signaling and the RTP packet by matching the acquired SIP signaling and RTP packet data according to timestamp information and payload type;

[0117] a call record data acquisition unit configured to acquire SIP signaling corresponding to a target call record and RTP packet data corresponding to the SIP signaling according to the correspondence between the SIP signaling and the RTP packet.

[0118] Optionally, the acquired SIP signaling comprises timestamp and session description protocol (SDP) information, and the acquired RTP packet data comprises sequence number, timestamp, synchronization source (SSRC) and payload type.

[0119] Optionally, the first and last packet identification module 12 comprises:

[0120] a connection time point determination unit configured to determine a time point of a first message in the SIP signaling as an indicator of a voice connection identifier as a voice connection time point;

[0121] a first packet identification unit configured to identify a first RTP packet after the voice connection time point in the RTP packet data as the first packet;

[0122] an end time point determination unit configured to determine a time point of a second message in the SIP signaling as an indicator of a voice call end identifier as a voice end time point;

[0123] a tail packet identification unit configured to identify a last RTP packet before the voice end time point in the RTP packet data as the tail packet.

[0124] Optionally, the voice quality evaluation module 14 comprises:

[0125] a first calculation unit configured to calculate a first time interval between any two continuous RTP packets in the valid RTP packets;

[0126] a third calculation unit configured to calculate a second time interval between the tail packet and the voice call end identifier;

[0127] an evaluation and judgment unit configured to evaluate the voice quality of the target call record according to the first time interval and the second time interval, to obtain an evaluation result of the voice quality.

[0128] Optionally, the evaluation and judgment unit is specifically configured to:

[0129] If any one of the first time intervals is greater than a preset first threshold or the second time interval is greater than a preset second threshold, it is determined that the evaluation result of the voice quality is that there is voice quality difference.

[0130] Optionally, the first threshold is 500 ms, and the second threshold is 1000 ms.

[0131] Embodiment 3

[0132] Reference Figure 7 The embodiment provides a voice quality evaluation device, which comprises a memory 21 and a processor 22. The memory 21 stores a computer program. The processor 22 is configured to run the computer program to execute the voice quality evaluation method in the embodiment 1.

[0133] The memory 21 is connected with the processor 22. The memory 21 can be a flash memory or a read-only memory or other memories. The processor 22 can be a central processing unit or a single-chip microcomputer.

[0134] Embodiment 4

[0135] The embodiment provides a computer readable storage medium, which stores a computer program. The computer program is executed by a processor to implement the voice quality evaluation method in the above-mentioned embodiment 1.

[0136] The computer-readable storage medium includes volatile or non-volatile, removable or non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, computer program modules or other data. The computer-readable storage medium includes, but is not limited to, RAM (Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable read only memory), flash memory or other memory technology, CD-ROM (Compact Disc Read-Only Memory), digital versatile disc (DVD) or other optical disc storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer.

[0137] In summary, the voice quality evaluation method, device and readable storage medium provided by the embodiments of the present application first acquire the session initiation protocol (SIP) signaling corresponding to the target call record and the real-time transport protocol (RTP) packet data corresponding to the SIP signaling; then identify the first packet and the last packet of the RTP packet according to the SIP signaling and the RTP packet data; then identify the first packet, the last packet and all the RTP packets between the first packet and the last packet in the RTP packet data as valid RTP packets; finally, evaluate the voice quality of the target call record according to the time interval between the valid RTP packets and the time interval between the last packet and the voice call end identifier, and obtain the evaluation result of the voice quality. The present application ensures that the voice quality evaluation is based on only the valid RTP packet data by prioritizing the validity judgment of the RTP packet. At the same time, the accuracy and reliability of the voice quality evaluation are significantly improved by comprehensively analyzing the time interval of the valid RTP packet and the time interval between the last packet and the voice call end identifier, so that the evaluation result is closer to the actual call experience. The existing voice quality evaluation method only evaluates the voice quality according to the relationship between the RTP packets, but ignores the validity of the RTP packets and the process judgment of the call start and end stage, thereby resulting in inaccurate evaluation results.

[0138] It can be understood that the above embodiments are only exemplary embodiments adopted for illustrating the principles of the present application, and the present application is not limited thereto. Various modifications and improvements can be made by those of ordinary skill in the art without departing from the spirit and essence of the present application, and these modifications and improvements are also considered to be within the protection scope of the present application.

Claims

1. A method of speech quality assessment, characterized by, The method comprises: acquiring session initiation protocol (SIP) signaling corresponding to a target call sheet and real-time transport protocol (RTP) packet data corresponding to the SIP signaling; identifying a first packet and a last packet of the RTP packet according to the SIP signaling and the RTP packet data; identifying the first packet, the last packet and all RTP packets between the first packet and the last packet in the RTP packet data as valid RTP packets; evaluating voice quality of the target call sheet according to a time interval between the valid RTP packets and a time interval between the last packet and a voice call end identifier, to obtain an evaluation result of the voice quality; the identifying of the first packet and the last packet of the RTP packet according to the SIP signaling and the RTP packet data specifically comprises: determining a time point of a first message in the SIP signaling as a voice on time point, the first message being an identifier indicating voice on; identifying a first RTP packet after the voice on time point in the RTP packet data as the first packet; determining a time point of a second message in the SIP signaling as a voice off time point, the second message being an identifier indicating voice off; identifying a last RTP packet before the voice off time point in the RTP packet data as the last packet.

2. The method of claim 1, wherein, the acquiring of the SIP signaling corresponding to the target call sheet and the RTP packet data corresponding to the SIP signaling specifically comprises: collecting SIP signaling and RTP packet data generated by a user during voice call; matching the collected SIP signaling and RTP packet data according to time stamp information and payload type, to obtain a corresponding relationship between the SIP signaling and the RTP packet; acquiring the SIP signaling corresponding to the target call sheet and the RTP packet data corresponding to the SIP signaling according to the corresponding relationship between the SIP signaling and the RTP packet.

3. The method of claim 2, wherein, The collected SIP signaling comprises time stamp and session description protocol (SDP) information, and the collected RTP packet data comprises sequence number, time stamp, synchronization source (SSRC) and payload type.

4. The method of claim 1, wherein, the evaluating of the voice quality of the target call sheet according to the time interval between the valid RTP packets and the time interval between the last packet and the voice call end identifier, to obtain the evaluation result of the voice quality, specifically comprises: calculating a first time interval between any two continuous RTP packets in the valid RTP packets; calculating a second time interval between the last packet and the voice call end identifier; evaluating the voice quality of the target call sheet according to the first time interval and the second time interval, to obtain the evaluation result of the voice quality.

5. The method of claim 4, wherein, the evaluating of the voice quality of the target call sheet according to the first time interval and the second time interval, to obtain the evaluation result of the voice quality, specifically comprises: if any one of the first time intervals is greater than a preset first threshold or the second time interval is greater than a preset second threshold, determining that the evaluation result of the voice quality is that there is voice quality difference.

6. The method of claim 5, wherein, The first threshold is 500 ms, and the second threshold is 1000 ms.

7. A speech quality assessment apparatus characterized by comprising: the device comprises: The call record data acquisition module is configured to acquire session initiation protocol (SIP) signaling corresponding to a target call record and real-time transport protocol (RTP) packet data corresponding to the SIP signaling. The head and tail packet identification module is connected to the call record data acquisition module and configured to identify a head packet and a tail packet of the RTP packet according to the SIP signaling and the RTP packet data. The valid RTP packet identification module is connected to the head and tail packet identification module and configured to identify the head packet, the tail packet and all RTP packets between the head packet and the tail packet in the RTP packet data as valid RTP packets. The voice quality evaluation module is connected to the valid RTP packet identification module and configured to evaluate voice quality of the target call record according to a time interval between the valid RTP packets and a time interval between the tail packet and a voice call end identifier, and obtain an evaluation result of the voice quality. The head and tail packet identification module includes: The connection time point determination unit is configured to determine a time point of a first message indicating a voice connection identifier in the SIP signaling as a voice connection time point. The head packet identification unit is configured to identify a first RTP packet after the voice connection time point in the RTP packet data as the head packet. The end time point determination unit is configured to determine a time point of a second message indicating a voice call end identifier in the SIP signaling as a voice end time point. The tail packet identification unit is configured to identify a last RTP packet before the voice end time point in the RTP packet data as the tail packet.

8. A speech quality assessment apparatus characterized by comprising: The computer readable storage medium stores a computer program, and the computer program is executed by the processor to implement the voice quality evaluation method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is executed by the processor to implement the voice quality evaluation method according to any one of claims 1-6.

Images