Voice call quality monitoring method, device, storage medium and program product

Abstract

The present disclosure discloses a method for monitoring the voice call quality of a media address, which includes: obtaining a call record generated by an outbound call system recording a target voice call, wherein the call record includes a call quality score, human voice recognition information, abnormal data packet information, and a media address of the target voice call, and the human voice recognition information is generated by identifying the target voice call through automatic voice recognition technology; obtaining status type information of the media address according to the call quality score, human voice recognition information, and abnormal data packet information of the target voice call; and controlling the call routing of the media address according to the status type information. The method of the present disclosure can monitor network anomalies of outbound voice calls, distinguish between business anomalies and network fluctuation anomalies of media addresses, select media addresses with good network conditions for outbound calls, optimize the voice call quality of the outbound call system, and improve user experience. The present disclosure also provides corresponding devices, computer storage media, and program products.

Description

Technical Field

[0001] The present disclosure relates to the field of communication technology, and in particular to a method for monitoring voice call quality. In addition, the present disclosure also relates to a related device, a storage medium, and a computer program product. Background Art

[0002] Currently, the Internet-based intelligent voice outbound calling service has been widely used in customer service, sales promotion, product marketing and other fields. However, the outbound calling service is also facing the adverse effects of network anomalies, which will reduce the accuracy of the outbound calling system in recognizing the user's voice, causing the outbound calling system to misunderstand or fail to understand the user's intention, thus reducing the user experience. Therefore, it is necessary to monitor and optimize the voice call quality of the outbound calling service.

[0003] In this regard, it has been proposed to use a connection protocol to monitor the delay and packet loss rate of a certain number of outbound call system server addresses, thereby monitoring the network quality.

[0004] However, this solution is not applicable to intelligent voice outbound calling services that use connectionless protocols, and the solution cannot distinguish whether network anomalies are caused by basic network problems or operator underlying service problems.

[0005] In view of this, there is an urgent need for an improved solution for voice call quality monitoring for outbound calling services.

[0006] The content of this background technology description is only for facilitating understanding of the relevant technology in this field and is not regarded as an admission of the prior art. Summary of the invention

[0007] Accordingly, the present invention provides a method and apparatus for monitoring voice call quality, and related storage media and program products, which at least partially solve the above-mentioned problems.

[0008] In a first aspect, an embodiment of the present disclosure provides a method for monitoring voice call quality of a media address, which includes:

[0009] Obtaining a call record generated by an outbound call system recording a target voice call, wherein the call record includes a call quality score, voice recognition information, abnormal data packet information, and a media address of the target voice call, and the voice recognition information is generated by recognizing the target voice call through an automatic voice recognition technology;

[0010] Acquire the status type information of the media address according to the call quality score, voice recognition information and abnormal data packet information of the target voice call;

[0011] The call routing of the media address is controlled according to the status type information.

[0012] Optionally, the acquiring the status type information of the media address includes:

[0013] When the call quality score of the target voice call is less than the network anomaly score threshold, the status type information of the media address is set to network anomaly.

[0014] Optionally, the voice recognition information of the target voice call includes a voice time period and a non-voice time period obtained by identifying the target voice call, and the abnormal data packet information of the target voice call includes the total amount of abnormal data packets, the time point of the abnormal data packets and the total amount of data packets of the target voice call;

[0015] The acquiring of the state type information of the media address further includes:

[0016] According to the human voice time period, the non-human voice time period, the time point of the abnormal data packet and the total number of abnormal data packets, the proportion of abnormal data packets in the human voice time period is obtained;

[0017] Obtaining a proportion of abnormal data packets according to the total amount of abnormal data packets and the total amount of data packets;

[0018] When the call quality score of the target voice call is greater than or equal to a network anomaly score threshold, the status type information of the media address is determined according to the proportion of abnormal data packets in the human voice time period of the target voice call and the proportion of abnormal data packets in the target voice call.

[0019] Optionally, the determining the state type information of the media address according to the proportion of abnormal data packets in the human voice time period in the target voice call and the proportion of abnormal data packets in the target voice call includes:

[0020] When the proportion of abnormal data packets in the voice time period is greater than or equal to the voice data packet abnormality threshold, and the proportion of abnormal data packets is greater than or equal to the business abnormality threshold, the status type information of the media address is set to business-type fluctuation abnormality.

[0021] Optionally, the determining the state type information of the media address according to the proportion of abnormal data packets in the human voice time period in the target voice call and the proportion of abnormal data packets in the target voice call further includes:

[0022] When the proportion of abnormal data packets in the voice time period is less than the voice data packet abnormality threshold, and the proportion of abnormal data packets is greater than or equal to the network abnormality threshold, the status type information of the media address is set to network type fluctuation abnormality.

[0023] Optionally, the determining the state type information of the media address according to the proportion of abnormal data packets in the human voice time period in the target voice call and the proportion of abnormal data packets in the target voice call further includes:

[0024] When the proportion of abnormal data packets in the voice time period is greater than or equal to the voice data packet abnormality threshold and the proportion of abnormal data packets is less than the service abnormality threshold, the state type information of the media address is set to a normal state.

[0025] Optionally, the determining the state type information of the media address according to the proportion of abnormal data packets in the human voice time period in the target voice call and the proportion of abnormal data packets in the target voice call further includes:

[0026] When the proportion of abnormal data packets in the human voice time period is less than the human voice data packet abnormality threshold and the proportion of abnormal data packets is less than the network abnormality threshold, the state type information of the media address is set to a normal state.

[0027] Optionally, controlling the call routing of the media address according to the status type information includes:

[0028] If the number of voice calls associated with the media address is greater than or equal to the monitoring threshold within the preset time period, the number of occurrences of various status type information of the media address within the preset time period is obtained;

[0029] The call routing of the media address is controlled according to the number of occurrences of various status type information of the media address, wherein the status type information includes network anomaly, business type fluctuation anomaly, network type fluctuation anomaly and normal status.

[0030] Optionally, controlling the call routing of the media address according to the number of occurrences of various status type information of the media address includes:

[0031] If the number of occurrences of any one or more of the various status type information is greater than or equal to the corresponding threshold or the corresponding preset proportion threshold, an alarm message is generated and the routing of the media address is stopped.

[0032] Optionally, if the number of occurrences of any one or more of the various status type information is greater than or equal to a corresponding threshold or a corresponding preset proportion threshold, an alarm message is generated and routing of the media address is stopped, including:

[0033] When the proportion of the number of network anomalies within the preset time period is greater than a first preset proportion threshold, an alarm message is generated and the routing of the media address is stopped, wherein the proportion of the number of network anomalies is obtained by the total number of various status type information within the preset time period and the number of network anomalies;

[0034] When the proportion of the number of abnormal business type fluctuations is greater than a second preset proportion threshold, an alarm message is generated and routing of the media address is stopped, wherein the proportion of the number of abnormal business type fluctuations is obtained by the total number and the number of abnormal business type fluctuations;

[0035] When the proportion of the number of network type fluctuation anomalies is greater than the third preset proportion threshold, an alarm message is generated and the routing of the media address is stopped, wherein the proportion of the number of network type fluctuation anomalies is obtained by the total number and the number of network type fluctuation anomalies.

[0036] Optionally, the method further comprises:

[0037] According to the routing control status of each of the media addresses, one of the available media addresses is selected from a plurality of available media addresses for outbound calling.

[0038] In a second aspect, an embodiment of the present disclosure provides a device for monitoring the voice call quality of a media address, which may include an acquisition module, a state type acquisition module and a control module, wherein:

[0039] The acquisition module is configured to acquire a call record generated by the outbound call system recording a target voice call, wherein the call record includes a call quality score, voice recognition information, abnormal data packet information and a media address of the target voice call, wherein the voice recognition information is generated by automatic voice recognition technology for the target voice call;

[0040] The state type acquisition module is configured to acquire the state type information of the media address according to the call quality score, voice recognition information and abnormal data packet information of the target voice call;

[0041] The control module is configured to control the call routing of the media address according to the status type information.

[0042] In a third aspect, an embodiment of the present disclosure provides a computer-readable storage medium having a computer program stored thereon, wherein when the program is executed by a processor, the method described in the first aspect is implemented.

[0043] In a fourth aspect, an embodiment of the present disclosure provides a computer program product, including a computer program, wherein when the computer program is executed by a processor, the method described in the first aspect is implemented.

[0044] The disclosed embodiment discloses a voice call quality monitoring method, which can obtain a call record generated by an outbound call system recording a target voice call, wherein the call record includes a call quality score, voice recognition information, abnormal data packet information and a media address of the target voice call; obtain the status type information of the media address according to the call quality score, voice recognition information and abnormal data packet information of the target voice call, and control the call routing of the media address according to the status type information. Through the method of the disclosed embodiment, the call quality of outbound voice calls can be monitored at the media transmission level, and can be widely applied to outbound call systems based on connection protocols / connectionless protocols. By comprehensively analyzing the call quality score, voice recognition information and abnormal data packet information, etc., it is possible to distinguish the specific network abnormality type of the media address, and select the media address with good network conditions for outbound calls, thereby optimizing the voice call quality of the outbound call system and improving the user experience.

[0045] Other optional features and technical effects of the embodiments of the present disclosure are partially described below, and partially can be understood by reading this document. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. The elements shown are not limited by the proportions shown in the accompanying drawings. The same or similar reference numerals in the accompanying drawings represent the same or similar elements, wherein:

[0047] Figure 1 An exemplary flow chart of a method for monitoring voice call quality according to an embodiment of the present disclosure is shown;

[0048] Figure 2 An exemplary flow chart of a method for monitoring voice call quality according to an embodiment of the present disclosure is shown;

[0049] Figure 3 A schematic diagram showing a method for monitoring voice call quality according to a specific embodiment of the present disclosure for obtaining the proportion of abnormal data packets in a human voice time period;

[0050] Figure 4 An exemplary flow chart of a method for monitoring voice call quality according to an embodiment of the present disclosure is shown;

[0051] Figure 5 An exemplary flow chart of a method for monitoring voice call quality according to an embodiment of the present disclosure is shown;

[0052] Figure 6 An exemplary flow chart of a method for monitoring voice call quality according to a specific embodiment of the present disclosure is shown;

[0053] Figure 7 An exemplary structural diagram of a voice call quality monitoring device according to an embodiment of the present disclosure is shown; and

[0054] Figure 8 An exemplary structural diagram of an electronic device capable of implementing the method according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the present disclosure more clear, the present disclosure is further described in detail below in conjunction with specific implementation methods and drawings. Here, the illustrative implementation methods and descriptions of the present disclosure are used to explain the present disclosure, but are not intended to limit the present disclosure.

[0056] In the embodiments of the present disclosure, reference Figure 1 , a method for monitoring the voice call quality of a media address is provided, which may at least include 110 to 130. In some embodiments, Figure 1 The method shown may be executed by a voice call quality monitoring system, which includes but is not limited to software / hardware that is independent of an outbound call system or integrated in an outbound call system, and the present disclosure does not impose any limitation on this.

[0057] 110: Obtain a call record generated by the outbound call system recording the target voice call, wherein the call record includes a call quality score, voice recognition information, abnormal data packet information, and a media address of the target voice call.

[0058] In an embodiment of the present disclosure, the outbound call system includes an intelligent voice outbound call system, which can make calls to preset recipients, recognize the recipient's voice and interact with the recipient in voice. In one example, the intelligent voice outbound call system can be configured with multiple media addresses for outbound calls, and the outbound call system can make voice calls with the recipients via these media addresses, wherein the media addresses include but are not limited to media IP addresses or ports, etc. In a specific example, the outbound call system is configured with multiple media IP addresses, and establishes, manages and terminates outbound calls based on media IP addresses based on the Session Initiation Protocol (SIP).

[0059] In some embodiments of the present disclosure, the transmission of the voice call is based on the Real-time Transport Protocol (RTP), and the audio of the voice call is transmitted in the form of data packets. The data packets may include audio data of an agreed time length and are sent continuously at agreed time intervals.

[0060] In the disclosed embodiment, the outbound call system can record the voice call with the callee and generate a call record, wherein the generated call record includes but is not limited to the call quality score of the voice call, voice recognition information, abnormal data packet information and media address, etc. In the disclosed embodiment, the information, data and signals involved are all authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0061] In some embodiments of the present disclosure, a voice call that is successfully connected by the outbound call system and communicates with the recipient may be referred to as a target voice call. However, the present application is not limited thereto. For example, in another embodiment, a voice call that is successfully connected or whose call time after connection is greater than a duration threshold may also be determined as a target voice call. The time threshold may be set according to actual conditions, and the present application does not limit this.

[0062] In some embodiments, the call record can be generated based on the call data of the voice call recorded by the outbound call system, wherein the call data includes but is not limited to the start time of the voice call, the call duration, the media addresses of the calling parties, the data packet information transmitted during the call, etc.

[0063] Accordingly, in some embodiments, before or at the time of step 110, the method may further include steps A1 to A2.

[0064] A1: Record the call data of voice calls.

[0065] In one example, the outbound call system may include a soft switch, and the outbound call system may use the soft switch to record the call data of the voice call, wherein the soft switch includes but is not limited to FreeSWITCH, Asterisk, OpenSIPS, etc. In a specific embodiment, the soft switch FreeSWITCH is used to record the call data of the voice call, such as recording a call detail record (CDR for short), and the CDR may include a media IP address (Media IP), a media port (Media Port), a call start timestamp (Start Time), a call end timestamp (End Time), a call duration (Duration), codec information (Codec), RTP statistics (RTP Statistics), etc. The CDR can be used to further generate call records.

[0066] A2: Generate call records based on the call data of the voice call.

[0067] As mentioned above, the call record includes but is not limited to the call quality score of the voice call, voice recognition information, abnormal data packet information and media address, etc.

[0068] In one example, the call quality score can be used to evaluate the quality of the voice call as a whole, and the score can be generated based on a preset scoring standard and based on the call data of the voice call recorded by the outbound call system. In one example, the preset scoring standard includes but is not limited to the voice quality scoring standard of the E-Model based on the transmission network parameters, the mean opinion score (MOS) quality scoring standard based on subjective voice evaluation, etc.

[0069] In a specific embodiment, the outbound call system uses the E-Model voice quality scoring standard, and the soft switch determines the basic signal-to-noise ratio, end-to-end delay, codec factors, and packet loss rate of the target voice call based on the codec information (Codec) and RTP statistical information of the target voice call in the recorded CDR, and determines the call quality score R based on the above information.

[0070] In the embodiment of the present disclosure, the abnormal data packet information may include the total amount of abnormal data packets, the time point of the abnormal data packets, and the total amount of data packets of the target voice call.

[0071] In one example, the soft switch of the outbound call system can determine the total amount of abnormal data packets, abnormal data packet time points, etc. during the target voice call based on the RTP statistical information in the CDR. The abnormal data packets may include lost data packets and delayed data packets during the target voice call, wherein the lost data packets include data packets that are not successfully transmitted, and the delayed data packets include data packets that are not sent / received at the agreed time interval.

[0072] In a specific embodiment, the call audio data packet of the target voice call is transmitted between the outbound call system media IP1 and the client IP2, wherein the call audio is collected at a sampling rate of 8kHz and encapsulated into an RTP data packet through voice coding. The RTP data packet is sent bidirectionally and continuously between IP1 and IP2 at an agreed time interval of 20ms, and each RTP data packet has a sequence number. The soft switch SWITCH determines that three consecutive data packets from IP2 are received according to the RTP statistical information in the CDR: "Data packet_100, receiving time_10:00:00. 000, IP2", "packet_102, receiving time_10:00:00.040", IP2, "packet_103, receiving time_10:00:00.075, IP2", wherein the sequence numbers of the adjacent data packets_100 and data packets_102 are not continuous, then data packet 101 is determined to be an abnormal data packet (lost data packet); although the sequence numbers of the adjacent data packets_102 and data packets_103 are continuous, the receiving time interval is 35ms, which does not meet the agreed time interval of 20ms, then data packet 103 is determined to be an abnormal data packet (delayed data packet). However, the present disclosure is not limited to this. People can imagine that the abnormal data packet information can also be obtained based on the information of the received RTP data packet returned by the client IP2, and these situations fall within the scope of the present disclosure.

[0073] In one example, the voice recognition information may be generated by the outbound call system through audio recognition of the target voice call, wherein the voice recognition information may include the voice time period and the non-voice time period obtained by recognizing the target voice call. In one example, the outbound call system may use ASR (Automatic Speech Recognition) technology to separate the voice of the target voice call audio and determine the voice time period and the non-voice time period.

[0074] In some embodiments of the present disclosure, the outbound calling system may selectively generate voice recognition information based on the call quality score of the target voice call, which will be further described below.

[0075] 120: Obtain the status type information of the media address according to the call quality score of the target voice call, the human voice recognition information and the abnormal data packet information.

[0076] In an embodiment of the present disclosure, the status type information may include network anomalies, business-type fluctuation anomalies and network-type fluctuation anomalies, wherein the network anomalies include serious network problems caused by network infrastructure failures; the business-type fluctuations include call quality fluctuations caused by communication intermediary services, wherein the communication intermediaries include but are not limited to communication operators and line operators, etc., and the communication intermediary services include but are not limited to call forwarding, call waiting, communication quality inspection services provided by communication operators, and transit nodes, communication data routing, transcoding processing, etc. provided by line operators; the network-type fluctuation anomalies include call quality fluctuations caused by random fluctuations or instability occurring during network transmission.

[0077] In some embodiments of the present disclosure, reference Figure 2 , the 120 may include 121 to 124:

[0078] 121 (unidentified): In response to the call quality score of the target voice call being less than the network anomaly score threshold, the status type information of the media address is set to network anomaly.

[0079] In the embodiment of the present disclosure, in combination with the description in reference 110, the call quality score can be determined according to a preset scoring standard, and a network anomaly score threshold corresponding to the scoring standard can be set.

[0080] In this embodiment, when the call quality score of the target voice call is less than the network anomaly score threshold, it indicates that the target voice call has a serious network problem, and the state type information of the media address corresponding to the target voice call is set to network anomaly. As an explanation and not limitation, the network problem includes but is not limited to packet loss, delay, etc.

[0081] In a specific embodiment, the outbound call system uses the E-Model voice quality scoring standard to determine the call quality score R of the target voice call, where R ranges from 0 to 100, and the network anomaly score threshold is set to 90. The monitoring tool of the outbound call system obtains the call quality score R from the soft switch. When R is less than 90, the monitoring tool sets the status type information of the media address corresponding to the target voice call to network anomaly.

[0082] Continuing with the above example, in a further specific embodiment of the present disclosure, the E-Model voice quality scoring standard and the MOS voice quality scoring standard may also be used in combination, wherein the R value determined by the E-Model voice quality scoring standard ranges from 0 to 100, and the MOS score determined by the MOS voice quality scoring standard ranges from 1 to 5. Based on the following table, the R value may be mapped to the MOS score, and according to the MOS scoring standard, the network anomaly score threshold may be determined to be 4, and when the MOS score corresponding to the target voice call is less than 4, the status type information of the media address corresponding to the target voice call is set to network anomaly.

[0083] Table 1 R value and MOS score evaluation criteria

[0084]

[0085] The above method of the present disclosure evaluates the voice call quality of the media address as a whole, and screens out the media addresses with serious network anomalies. However, for voice calls with a call quality score greater than or equal to the network anomaly score threshold, the corresponding media addresses may still have certain network anomalies, which can be further distinguished according to the method of the following embodiment of the present disclosure.

[0086] In another embodiment, as described above, the outbound call system can selectively generate human voice recognition information based on the call quality score of the target voice call. In one example, when the call quality score of the target voice call is greater than or equal to the network anomaly score threshold, the monitoring tool of the outbound call system will generate human voice recognition information based on the audio of the target voice call, otherwise the audio of the target voice call will not be obtained. As an explanation and not a limitation, when the call quality score of the target voice call is less than the network anomaly score threshold, in other words, when it has been determined that the status type information of the media address corresponding to the target voice call is a network anomaly, the steps of obtaining the target voice call audio and identifying it can be omitted, which is conducive to saving the operating load of the monitoring tool of the outbound call system and improving the efficiency of voice quality monitoring.

[0087] 122: According to the human voice time period, the non-human voice time period, the time point of the abnormal data packet and the total number of abnormal data packets, the proportion of abnormal data packets in the human voice time period is obtained.

[0088] In the embodiment of the present disclosure, the abnormal data packet ratio in the human voice time period includes the ratio of the number of abnormal data packets falling in the human voice time period to the total number of abnormal data packets. In this embodiment, the content of how to determine the abnormal data packets can refer to the description in the aforementioned 110, which will not be repeated here.

[0089] In some embodiments of the present disclosure, the 122 may include: obtaining the number of abnormal data packets in the human voice time period according to the time point of each abnormal data packet, and calculating the proportion of abnormal data packets in the human voice time period according to the number of abnormal data packets in the human voice time period and the total number of abnormal data packets.

[0090] In a specific embodiment, if Figure 3 As shown, the audio of voice call A starts from t0 and ends at t9. After ASR performs voice separation, the voice time periods P1, P2, P3, P4, and P5 are determined. According to the RTP statistical information in the CDR, it is determined that the total number of abnormal RTP data packets in voice call A is 8, and the corresponding time points are t1 to t8, among which the number of data packets falling into the voice time period is 5, and the corresponding time points are t1, t3, t4, t5, and t8, respectively. It is calculated that the proportion of abnormal data packets in the voice time period = 5 / 8 × 100% = 62.5%.

[0091] 123: Gets the abnormal data packet ratio based on the total number of abnormal data packets and the total number of data packets.

[0092] In this embodiment, the outbound call system can determine the total amount of abnormal RTP packets and the total amount of transmitted RTP packets based on the RTP statistical information in the CDR, and determine the abnormal packet ratio = the total amount of abnormal RTP packets / the total amount of RTP packets. Figure 3 As shown, the audio of voice call A starts from t0 and ends at t9. According to the RTP statistical information in the CDR, it is determined that the total number of RTP data packets transmitted in voice call A is 160, and the total number of abnormal RTP data packets is 8. The corresponding time points are t1 to t8, and the calculated abnormal data packet ratio = total number of abnormal RTP data packets / total number of RTP data packets = 8 / 160×100%=5%.

[0093] Here, it can be understood that the above steps 122 and 123 are only functional distinctions between two steps, and do not imply that the two steps are arranged in a given order. 122 and 123 can be executed sequentially in any order or in parallel.

[0094] 124: In response to the call quality score of the target voice call being greater than or equal to the network anomaly score threshold, determine the status type information of the media address according to the proportion of abnormal data packets in the human voice time period of the target voice call and the proportion of abnormal data packets in the target voice call.

[0095] In the embodiment of the present disclosure, for voice calls whose call quality score is greater than or equal to the network anomaly score threshold, the cause of the network anomaly can be further subdivided by combining the abnormal data packet ratio and the abnormal data packet ratio in the human voice time period.

[0096] Accordingly, in some embodiments of the present disclosure, the 124 may include B1 and B2.

[0097] B1: in response to the fact that the proportion of abnormal data packets in the human voice time period is greater than or equal to the human voice data packet abnormality threshold, and the proportion of abnormal data packets is greater than or equal to the service abnormality threshold, setting the state type information of the media address to service type fluctuation abnormality.

[0098] B2: In response to the fact that the proportion of abnormal data packets in the voice time period is greater than or equal to the voice data packet abnormality threshold, and the proportion of abnormal data packets is less than the service abnormality threshold, setting the state type information of the media address to a normal state.

[0099] In some embodiments of the present disclosure, for example, in the above B1 and B2, when the proportion of abnormal data packets in the human voice time period is greater than or equal to the human voice data packet abnormality threshold, it means that the network fluctuation mainly occurs in the time period when the user is speaking, rather than being evenly distributed throughout the voice call. Such network fluctuations may be caused by performance problems of the receiving device and / or problems in the business of the communication intermediary. This can be judged in combination with the proportion of abnormal data packets: when the proportion of abnormal data packets is greater than or equal to the business abnormality threshold, the status type information of the media address is set to a business-type fluctuation abnormality; otherwise, the status type information of the media address is set to a normal state.

[0100] In a specific embodiment of the present disclosure, the voice data packet abnormality threshold may be set to 80%, and the service abnormality threshold may be set to 2%. However, it is understandable that in some embodiments, a higher / lower abnormality threshold may be set.

[0101] In some embodiments of the present disclosure, the 124 may also include B3 and B4.

[0102] B3: In response to the fact that the proportion of abnormal data packets in the human voice time period is less than the human voice data packet abnormality threshold, and the proportion of abnormal data packets is greater than or equal to the network abnormality threshold, the state type information of the media address is set to network type fluctuation abnormality.

[0103] B4: In response to the fact that the proportion of abnormal data packets in the human voice time period is less than the human voice data packet abnormality threshold, and the proportion of abnormal data packets is less than the network abnormality threshold, the state type information of the media address is set to a normal state.

[0104] In some embodiments of the present disclosure, for example, in B3 and B4 above, when the proportion of abnormal data packets in the human voice time period is less than the human voice data packet abnormality threshold, it means that the network fluctuation is evenly distributed in the entire voice call. This network fluctuation may originate from random fluctuations that occur during network transmission. It can be judged based on the proportion of abnormal data packets: when the proportion of abnormal data packets is greater than or equal to the network abnormality threshold, the status type information of the media address is set to a network type fluctuation abnormality; otherwise, the status type information of the media address is set to a normal state.

[0105] In a specific embodiment of the present disclosure, the abnormal threshold of the human voice data packet can be set to 80%, and the abnormal network fluctuation can be set to 5%. However, it can be understood that in some embodiments, a higher / lower abnormal threshold can be set.

[0106] Here, it can be understood that the above steps B1, B2, B3, and B4 are only distinguished in function, and do not imply that these steps are performed in a given order.

[0107] As an explanation but not limitation, in the intelligent voice outbound call service, the possibility of abnormal business fluctuations caused by the influence of the operator is low, but the abnormality will have an impact on the outbound call service as a whole, while network-type fluctuation abnormalities usually appear in a certain time period or on certain media addresses. Therefore, in some embodiments of the present disclosure, for example, in the above B1, B2, B3, and B4, optionally, the business abnormality threshold is less than or equal to the network abnormality threshold. In a specific embodiment of the present disclosure, the business abnormality threshold is set to 2% and the network abnormality threshold is set to 5%.

[0108] 130: Control call routing of the media address according to the status type information.

[0109] In the disclosed embodiment, the outbound call system continuously monitors the voice calls via the media address, and stores the information of the voice calls and the media address status type information in the status type history record of the media address, which includes but is not limited to the media address, the start and end timestamps of the voice calls via the media address, the determined status type, etc. In a specific embodiment, a record in the status type history record of the media IP address A is "IP address_111.111.1.1, call_2024-01-01_11:00:00-11:05:00, status type_normal".

[0110] In some embodiments of the present disclosure, for example, in the above 130, outbound calls of at least part of the media addresses may be adjusted based on the determined status type information of at least part of the media addresses.

[0111] In some embodiments of the present disclosure, reference Figure 4, the 130 may include at least 131 to 132.

[0112] 131: within a preset time period, if the number of voice calls associated with the media address is greater than or equal to the monitoring threshold, the number of occurrences of various status type information of the media address within the preset time period is obtained.

[0113] In some embodiments of the present disclosure, within a preset time period, the outbound call system can determine the number of voice calls made through a media address in the outbound call system, and the number of voice calls includes but is not limited to the number of outbound calls made by the media address and the number of voice calls received by the media address. When the outbound call system determines that the number of voice calls is greater than or equal to a monitoring threshold, the outbound call system will retrieve the status type history record of the media address, and obtain the number of times various status type information of the media address appears within the preset time period based on the status type history record.

[0114] In a specific embodiment, the monitoring threshold is set to 10, the preset time length is 1 hour, and the monitoring tool of the outbound call system determines that the number of outbound calls made by the media IP address 111.111.1.1 within 1 hour is equal to 10. The monitoring tool reads the status type history record of the media IP address, and determines that the number of times various status type information appears in the media IP address in the past hour is: network abnormality: 0, business-type fluctuation abnormality: 1, network-type fluctuation abnormality: 1, normal: 8.

[0115] 132: Control the call routing of the media address according to the number of occurrences of various status type information of the media address.

[0116] In this embodiment, the status type information includes network abnormality, business type fluctuation abnormality, network type fluctuation abnormality and normal status.

[0117] In some embodiments of the present disclosure, the 132 may include: if the number of occurrences of any one or more of the various status type information is greater than or equal to the corresponding threshold or the corresponding preset proportion threshold, an alarm message is generated and the routing of the media address is stopped.

[0118] In this embodiment, when a certain media address has an alarm message, the routing of the media address can be stopped, that is, the media address is no longer used to establish any voice call. In a specific example, the outbound call system can maintain a list of available media addresses, and when a certain media address has an alarm message, the address can be removed from the list of available media addresses.

[0119] In this embodiment, if the number of occurrences of any one or more of the various status type information is greater than or equal to the corresponding threshold or the corresponding preset proportion threshold, generating an alarm message and stopping the routing of the media address may include at least C1 to C3.

[0120] C1: In response to the fact that the proportion of the number of network anomalies within the preset time period is greater than a first preset proportion threshold, an alarm message is generated and routing of the media address is stopped.

[0121] In this embodiment, the network anomaly count ratio is obtained by the total number of various status type information within the preset time length and the network anomaly count, that is, the network anomaly count ratio = the number of network anomalies within the preset time length / the total number of various status type information within the preset time length.

[0122] In a specific embodiment, the first preset percentage threshold is set to 50%, and the monitoring tool of the outbound call system determines that the number of outbound calls made by the media IP address 111.111.1.1 within 1 hour is equal to 10. The monitoring tool begins to read the status type history record of the media IP address, and determines that the number of times various status type information appears in the media IP address in the past hour is: network abnormality: 6, business type fluctuation abnormality: 1, network type fluctuation abnormality: 0, normal: 3, and the calculated network abnormality percentage = 6 / 10×100% = 60%>50%. At this time, an alarm message will be generated and the routing of the IP address 111.111.1.1 will be stopped.

[0123] C2: In response to the proportion of the number of abnormal business fluctuations being greater than a second preset proportion threshold, an alarm message is generated and routing of the media address is stopped.

[0124] In this embodiment, the business type fluctuation abnormal number ratio is obtained by the total number and the business type fluctuation abnormal number, that is, the business type fluctuation abnormal number ratio = the business type fluctuation abnormal number within the preset time length / the total number of various status type information within the preset time length. In a specific embodiment, the second preset ratio threshold can be set to 20%.

[0125] C3: In response to the fact that the proportion of the number of network-type fluctuation anomalies is greater than a third preset proportion threshold, an alarm message is generated and routing of the media address is stopped.

[0126] In this embodiment, the network type fluctuation abnormal number ratio is obtained by the total number and the network type fluctuation abnormal number, that is, the network type fluctuation abnormal number ratio = the network type fluctuation abnormal number within the preset time length / the total number of various status type information within the preset time length. In a specific embodiment, the third preset ratio threshold can be set to 30%.

[0127] It is understandable that the numerical value of the percentage threshold in the specific embodiment of the present disclosure is only an example, and those skilled in the art can adjust the percentage threshold as needed, and the present disclosure does not limit this. The specific calculation process of the abnormal number percentage in 1322 to 1323 in the above embodiment of the present disclosure is similar to the calculation process in the aforementioned 1321, and will not be repeated here.

[0128] In some embodiments of the present disclosure, for example, in 1321 to 1323, if any one or more of the number of network anomalies, business-type fluctuation anomalies, and network-type fluctuation anomalies of a certain media address in the past hour is greater than or equal to the corresponding threshold or the corresponding preset proportion threshold, an alarm message is generated and the routing of the media address is stopped. In some embodiments, if the number of network anomalies, business-type fluctuation anomalies, and network-type fluctuation anomalies of a certain media address in the past hour are all less than the corresponding threshold or the corresponding preset proportion threshold, no alarm message is generated and the routing of the media address is continued.

[0129] In some embodiments of the present disclosure, for a media address that has not generated an alarm message, the outbound call system may continue to monitor the media address, calculate the number of occurrences of various status type information within a preset time period before the latest voice call of the media address, and control the call routing of the media address.

[0130] In some other embodiments of the present disclosure, a monitoring time threshold may be set. For a media address that does not generate an alarm message, the outbound call system may continue to monitor the media address within the monitoring time threshold. If the media address does not generate an alarm message within the monitoring time threshold, and the number of voice calls associated with the media address within a preset time is less than the monitoring threshold, the monitoring of the media address may be stopped.

[0131] As mentioned above, the outbound call system can be configured with multiple media addresses for outbound calls. When or after the routing control status of at least some media addresses is determined according to the above method of the present invention, an alarm message is generated and the routing of media addresses with abnormal network status is stopped, a media address with a normal routing control status can be selected for outbound calls.

[0132] Accordingly, in some embodiments of the present disclosure, reference Figure 5 The voice call quality monitoring method may further include 140:

[0133] 140: Selecting an available media address from a plurality of available media addresses for outbound calling according to the routing control status of each media address.

[0134] In this embodiment, the available media addresses include media addresses whose routing control status is normal.

[0135] In some embodiments of the present disclosure, for example, in the above 140, the status index of each media address can be calculated / weighted based on at least one of the call quality score, voice recognition information and abnormal data packet information, and the available media address with the best status can be selected from multiple available media addresses for outbound calls according to preset sorting rules, such as ascending / descending sorting.

[0136] In one example, the average call quality scores of the most recent or agreed number of voice calls for multiple available media addresses may be calculated, the averages may be sorted, and the available media address with the highest average may be selected for outbound calls. However, it is to be understood that the above strategy is only used as an example and is not intended to limit the present disclosure. In other embodiments, more strategies may be used to randomly select an available media address from multiple available media addresses for outbound calls.

[0137] In a specific example, the outbound call system currently has two available media addresses IP1 and IP2. The outbound call system obtains the call quality scores Q of the last five voice calls from the historical records of each available media address. n , and the following data are obtained: IP1: Q1 = 91, Q2 = 92, Q3 = 90, Q4 = 93, Q5 = 92, IP2: Q1 = 90, Q2 = 92, Q3 = 93, Q4 = 95, Q5 = 95, and based on the above data, the average call quality score S of the available media addresses IP1 and IP2 is calculated respectively. n : For IP1: S1 = (91 + 92 + 90 + 93 + 92) / 5 = 91.6, for IP2: S2 = (90 + 92 + 93 + 95 + 95) / 5 = 93.75, since S2 > S1, the available media address IP2 is selected for outbound calls.

[0138] The method of the above-mentioned embodiment of the present disclosure can monitor the call quality of outbound voice calls at the media transmission level, and can be widely applicable to outbound call systems based on connection protocol / connectionless protocol. By comprehensively analyzing call quality scores, voice recognition information, and abnormal data packet information, etc., it can distinguish the specific network anomaly type of the media address to continuously control the routing of the media address, select the media address with good network conditions for outbound calls, optimize the voice call quality of the outbound call system, and improve the user experience.

[0139] The voice quality monitoring method of the present disclosure will be further described below in conjunction with a specific embodiment.

[0140] In a specific embodiment, reference Figure 6The voice quality monitoring method disclosed in the present invention is applied to an intelligent voice outbound call system. In this embodiment, the intelligent voice outbound call system is configured with 6 media addresses: IP1, IP2, IP3, IP4, IP5, and IP6, and these 6 media addresses are used to make voice calls for outbound calls. After each voice call, the outbound call system generates a call record for the voice call involved in each media address, wherein the call record includes the call quality score of the voice call, human voice recognition information, abnormal data packet information, and media address.

[0141] In this embodiment, the outbound call system monitors the voice quality of each media address based on the call record of the media address:

[0142] For IP1, after the end of IP1's most recent voice call T1, the outbound call system obtains the call record R1 generated by the voice call, queries it to obtain the call quality score (i.e., MOS value) Q1 corresponding to IP1, and compares Q1 with the predetermined call quality score threshold 4: Q1=3.8<4, indicating that the call quality of IP1's voice call is very poor. The outbound call system adds a new record in the status type history record of IP1: "IP1, call_2024-01-01_11:01:00-11:02:00, status type_network abnormality".

[0143] After updating the status type history record of IP1, the outbound call system obtains the records of the last 1 hour in the status type history record: Network abnormality: 1, Business type fluctuation abnormality: 0, Network type fluctuation abnormality: 1, Normal: 8, and the calculated network abnormality ratio is = 1 / 10×100% = 10% <50%, Business type fluctuation abnormality ratio = 0, Network type fluctuation abnormality ratio = 1 / 10×100% = 10% <30%. At this time, no alarm information is generated and the routing of media address IP1 continues.

[0144] For IP2, the outbound call system obtains the call record R2 generated by the voice call after the end of IP2's most recent voice call T2, and obtains the call quality score Q2 corresponding to IP2 from it, and compares Q2 with the predetermined call quality score threshold 4: Q2 = 4.1> 4, indicating that the call quality of IP2's voice call is acceptable. The outbound call system further obtains the human voice recognition information and abnormal data packet information of the voice call T2 from the call record R2, calculates the abnormal data packet ratio X2 in the human voice time period based on this information, and compares X2 with the predetermined call quality score threshold 80%: the total number of abnormal RTP data packets in the voice call T2 is 11, of which the number of data packets falling into the human voice time period is 9, and it is calculated that X2 = 9 / 11 × 100% = 81.8%> 80%. The outbound call system further calculates the abnormal data packet proportion Y3 based on the abnormal data packet information, and compares Y3 with the predetermined business type fluctuation abnormal threshold value of 2%: the total number of RTP data packets transmitted in the voice call T2 is 360, and the total number of abnormal RTP data packets is 11. It is calculated that Y3=11 / 360×100%=3.05%>2%. The outbound call system adds a new record in the status type history record of IP2: "IP2, call_2024-01-01_11:01:00-11:04:00, status type_business type fluctuation abnormality".

[0145] After updating the status type history record of IP2, the outbound call system obtains the records of the last hour in the status type history record: Network abnormality: 0, Business type fluctuation abnormality: 2, Network type fluctuation abnormality: 1, Normal: 9, and calculates that the network abnormality ratio = 0, the business type fluctuation abnormality ratio = 2 / 12×100% = 16.7% < 20%, the network type fluctuation abnormality ratio = 1 / 12×100% = 8.3% < 30%. At this time, no alarm information is generated and the routing of media address IP2 continues.

[0146] For IP3, the outbound call system obtains the call record R3 generated by the voice call after the end of IP3's most recent voice call T2, and obtains the call quality score Q3 corresponding to IP3 from it, and compares Q3 with the predetermined call quality score threshold 4: Q3 = 4.2> 4, indicating that the call quality of IP3's voice call is acceptable. The outbound call system further obtains the human voice recognition information and abnormal data packet information of the voice call T3 from the call record R3, calculates the abnormal data packet ratio X3 in the human voice time period based on this information, and compares X3 with the predetermined call quality score threshold 80%: the total number of abnormal RTP data packets in the voice call T3 is 32, of which the number of data packets falling into the human voice time period is 17, and it is calculated that X3 = 17 / 32 × 100% = 53.1% < 80%. The outbound call system further calculates the abnormal data packet proportion Y3 based on the abnormal data packet information, and compares Y3 with the predetermined network type fluctuation abnormal threshold of 5%: the total number of RTP data packets transmitted in the voice call T3 is 480, and the total number of abnormal RTP data packets is 32. It is calculated that Y3=32 / 480×100%=6.67%>5%. The outbound call system adds a new record in the status type history record of IP3: "IP3, call_2024-01-01_11:00:00-11:08:00, status type_network type fluctuation abnormality".

[0147] After updating the status type history record of IP3, the outbound call system obtains the records of the last hour in the status type history record: Network anomaly: 4, Business type fluctuation anomaly: 1, Network type fluctuation anomaly: 0, Normal: 2, and the calculated network anomaly ratio = 4 / 6×100% = 66.7%>50%, then an alarm message is generated and the routing of media address IP3 is stopped.

[0148] For IP4, similar to IP1, the outbound call system obtains the call record R4 generated by the voice call after the end of IP4’s most recent voice call T4, and queries it to obtain the call quality score Q4 corresponding to IP4 and compares Q4 with the predetermined call quality score threshold 4: Q4=4.5>4, indicating that the call quality of IP4’s voice call is good. The outbound call system adds a new record in the status type history record of IP4: "IP4, call_2024-01-01_10:00:00-10:05:00, status type_normal".

[0149] After updating the status type history record of IP4, the outbound call system obtains the records of the last hour in the status type history record: Network abnormality: 0, Business type fluctuation abnormality: 1, Network type fluctuation abnormality: 0, Normal: 9, and the calculated network abnormality ratio is 0, Business type fluctuation abnormality ratio = 1 / 10×100% = 10% < 20%, Network type fluctuation abnormality ratio = 0, then no alarm information is generated and the routing of media address IP4 continues.

[0150] For IP5, the outbound call system obtains the call record R5 generated by the voice call after the end of IP5's most recent voice call T5, and obtains the call quality score Q2 corresponding to IP5 from it and compares Q5 with the predetermined call quality score threshold 4: Q5 = 4.1> 4, indicating that the call quality of IP5's voice call is acceptable. The outbound call system further obtains the human voice recognition information and abnormal data packet information of the voice call T5 from the call record R5, calculates the abnormal data packet ratio X5 in the human voice time period based on this information, and compares X5 with the predetermined call quality score threshold 80%: the total number of abnormal RTP data packets in the voice call T5 is 8, of which the number of data packets falling into the human voice time period is 7, and it is calculated that X5 = 7 / 8 × 100% = 87.5%> 80%. The outbound call system further calculates the abnormal data packet proportion Y5 based on the abnormal data packet information, and compares Y5 with the predetermined business-type fluctuation abnormal threshold of 2%: the total number of RTP data packets transmitted in the voice call T5 is 360, and the total number of abnormal RTP data packets is 1, and it is calculated that Y5=7 / 360×100%=1.94%><2%. The outbound call system adds a new record in the status type history record of IP5: "IP5, call_2024-01-01_11:02:00-11:06:00, status type_normal".

[0151] After updating the status type history record of IP5, the outbound call system obtains the records of the last hour in the status type history record: Network abnormality: 0, Business type fluctuation abnormality: 2, Network type fluctuation abnormality: 1, Normal: 9, and calculates that the network abnormality ratio = 0, the business type fluctuation abnormality ratio = 2 / 12×100% = 16.7% < 20%, the network type fluctuation abnormality ratio = 1 / 12×100% = 8.3% < 30%. At this time, no alarm information is generated and the routing of media address IP5 continues.

[0152] For IP6, the outbound call system obtains the call record R6 generated by the voice call after the end of IP6's most recent voice call T6, and obtains the call quality score Q6 corresponding to IP6 from it and compares Q6 with the predetermined call quality score threshold 4: Q6 = 4.2> 4, indicating that the call quality of IP3's voice call is acceptable. The outbound call system further obtains the human voice recognition information and abnormal data packet information of the voice call T6 from the call record R3, calculates the abnormal data packet ratio X6 in the human voice time period based on this information, and compares X6 with the predetermined call quality score threshold 80%: the total number of abnormal RTP data packets in the voice call T6 is 22, of which the number of data packets falling into the human voice time period is 10, and it is calculated that X6 = 10 / 22 × 100% = 45.5% < 80%. The outbound call system further calculates the abnormal data packet proportion Y6 based on the abnormal data packet information, and compares Y6 with the predetermined network type fluctuation abnormal threshold value of 5%: the total number of RTP data packets transmitted in the voice call T6 is 480, and the total number of abnormal RTP data packets is 22. It is calculated that Y6=22 / 480×100%=4.58%<5%. The outbound call system adds a new record in the status type history record of IP6: "IP6, Call_2024-01-01_11:00:00-11:08:00, Status Type_Normal".

[0153] After updating the status type history record of IP6, the outbound call system obtains the records of the last hour in the status type history record: Network abnormality: 0, Business type fluctuation abnormality: 1, Network type fluctuation abnormality: 0, Normal: 6, and the calculated network abnormality ratio = 0, Business type fluctuation abnormality ratio = 1 / 7×100% = 14.3% < 20%, Network type fluctuation abnormality ratio = 0, then no alarm information is generated and the routing of media address IP5 continues.

[0154] The outbound call system continues to perform the monitoring process as described above, monitoring the voice call quality of the six media addresses IP1, IP2, IP3, IP4, IP5, and IP6.

[0155] When the outbound call system is preparing to establish an outbound voice call at 2024-01-01_12:00:00, the outbound call system detects that the available media addresses that have not generated alarm information in IP1, IP2, IP3, and IP4 are: IP1, IP2, IP4, IP5, and IP6. At this time, the outbound call system obtains the call quality scores Q of the most recent five voice calls of IP1, IP2, IP4, IP5, and IP6 respectively. n, and the following data are obtained: IP1: Q1=91, Q2=92, Q3=90, Q4=93, Q5=92, IP2: Q1=90, Q2=92, Q3=93, Q4=95, Q5=95, IP4: Q1=95, Q2=93, Q3=95, Q4=94, Q5=95, IP5: Q1=95, Q2=93, Q3=92, Q4=94, Q5=95, IP6: Q1=95, Q2=93, Q3=91, Q4=94, Q5=95, Based on the above data, the average call quality score S of the available media addresses IP1, IP2, IP4, IP5, and IP6 are calculated respectively. n : For IP1: S1 = (91+92+90+93+92) / 5 = 91.6, for IP2: S2 = (90+92+93+95+95) / 5 = 93.75, for IP2: S2 = (90+92+93+95+95) / 5 = 93.75, since S2>S1, for IP4: S4 = (95+93+95+94+95) / 5 = 94.4, for IP5: S4 = (95+93+92+94+95) / 5 = 93.8, for IP6: S4 = (95+93+91+94+95) / 5 = 93.6, since S4>S5>S2>S4>S1, so the available media address IP4 is selected for outbound call.

[0156] Accordingly, in some embodiments of the present disclosure, Figure 7 As shown, a device 700 for monitoring voice call quality of a media address is also provided, which may include an acquisition module 710, a state type acquisition module 720 and a control module 730, wherein:

[0157] The acquisition module configuration 710 is configured to acquire a call record generated by an outbound call system recording a target voice call, wherein the call record includes a call quality score, voice recognition information, abnormal data packet information and a media address of the target voice call, wherein the voice recognition information is generated for the target voice call through automatic speech recognition technology.

[0158] The state type acquisition module 720 is configured to acquire the state type information of the media address according to the call quality score, human voice recognition information and abnormal data packet information of the target voice call;

[0159] The control module 730 is configured to control the call routing of the media address according to the status type information.

[0160] In the embodiments of the present disclosure, the device may be combined with the features of the method for monitoring voice call quality in any embodiment, and vice versa, which will not be elaborated herein.

[0161] In an embodiment of the present disclosure, an electronic device is provided, including: a processor and a memory storing a computer program, wherein the processor is configured to execute the method for voice call quality monitoring of any embodiment of the present disclosure when running the computer program.

[0162] Figure 8 A schematic diagram of an electronic device 800 that can implement the method of the disclosed embodiment or implement the disclosed embodiment is shown, and in some embodiments, more or fewer electronic devices may be included than shown. In some embodiments, it can be implemented using a single or multiple electronic devices. In some embodiments, it can be implemented using cloud or distributed electronic devices.

[0163] like Figure 8 As shown, the electronic device 800 includes a central processing unit (CPU) 801, which can perform various appropriate operations and processes according to the programs and / or data stored in the read-only memory (ROM) 802 or the programs and / or data loaded from the storage part 808 to the random access memory (RAM) 803. CPU 801 can be a multi-core processor, or it can include multiple processors. In some embodiments, CPU 801 can include a general main processor and one or more special coprocessors, such as a graphics processing unit (GPU), a neural network processor (NPU), a digital signal processor (DSP), etc. In RAM 803, various programs and data required for the operation of the electronic device 800 are also stored. CPU 801, ROM 802 and RAM 803 are connected to each other via bus 804. Input / output (I / O) interface 805 is also connected to bus 804.

[0164] The processor and the memory are used together to execute the program stored in the memory. When the program is executed by the computer, the steps or functions of the method for monitoring voice call quality described in the above embodiments can be implemented.

[0165] The following components are connected to the I / O interface 805: an input section 806 including a keyboard, a mouse, a touch screen, etc.; an output section 807 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 808 including a hard disk, etc.; and a communication section 809 including a network interface card such as a LAN card, a modem, etc. The communication section 809 performs communication processing via a network such as the Internet. A drive 810 is also connected to the I / O interface 805 as needed. A removable medium 811, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 810 as needed, so that a computer program read therefrom is installed into the storage section 808 as needed. Figure 8 Only some components are schematically shown in the figure, which does not mean that the computer system 800 only includes Figure 8 Components shown.

[0166] In some embodiments, the electronic device 800 refers to a mobile terminal or a computer, including a mobile phone, a car terminal, a smart TV, etc. Taking a mobile phone as an example, the electronic device 800 also includes a display screen with a touch function, an external speaker, a gyroscope, a camera, a 4G / 5G antenna and other device modules.

[0167] The systems, devices, modules or units described in the above embodiments may be implemented by a computer or its associated components. The computer may be, for example, a mobile terminal, a smart phone, a personal computer, a laptop computer, a vehicle-mounted human-computer interaction device, a personal digital assistant, a media player, a navigation device, a game console, a tablet computer, a wearable device, a smart TV, an Internet of Things system, a smart home, an industrial computer, a server or a combination thereof.

[0168] Although not shown, in an embodiment of the present disclosure, a storage medium is provided, the storage medium storing a computer program, the computer program being configured to execute any method of the embodiments of the present disclosure when executed.

[0169] Although not shown, in an embodiment of the present disclosure, a computer program product is provided, including a computer program, wherein when the computer program is executed by a processor, the method described in any one of the embodiments of the present disclosure is implemented.

[0170] The storage media in the embodiments of the present disclosure include permanent and non-permanent, removable and non-removable items that can be implemented by any method or technology to store information. Examples of storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission medium that can be used to store information that can be accessed by a computing device.

[0171] The methods, programs, systems, devices, etc. of the embodiments of the present disclosure may be executed or implemented in a single or multiple networked computers, or may be practiced in a distributed computing environment. In the embodiments of the present specification, in these distributed computing environments, tasks may be performed by remote processing devices connected via a communication network.

[0172] Those skilled in the art should understand that the embodiments of the present specification may be provided as methods, systems or computer program products. Therefore, those skilled in the art may imagine that the functional modules / units or controllers and related method steps described in the above embodiments may be implemented in software, hardware or a combination of software / hardware.

[0173] Unless explicitly stated, the actions or steps of the methods, programs, and embodiments of the present disclosure do not have to be performed in a specific order and can still achieve the desired results. In some implementations, multitasking and parallel processing are also possible or may be advantageous.

[0174] In this article, multiple embodiments of the present disclosure are described, but for the sake of brevity, the description of each embodiment is not exhaustive, and the same or similar features or parts between the embodiments may be omitted. In this article, "one embodiment", "some embodiments", "example", "specific example", or "some examples" are meant to be applicable to at least one embodiment or example according to the present disclosure, rather than all embodiments. The above terms do not necessarily mean to refer to the same embodiment or example. In the absence of contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples.

[0175] The exemplary systems and methods of the present disclosure have been specifically shown and described with reference to the above embodiments, which are merely examples of the best modes for implementing the present systems and methods. It will be appreciated by those skilled in the art that various changes may be made to the embodiments of the systems and methods described herein when implementing the present systems and / or methods without departing from the spirit and scope of the present disclosure as defined in the appended claims.

Claims

1. A method for monitoring the voice call quality of a media address, characterized in that: include: Obtain a call record generated by an outbound call system recording a target voice call, wherein the call record includes a call quality score, voice recognition information, abnormal data packet information, and a media address of the target voice call, the voice recognition information is generated by recognizing the target voice call using an automatic voice recognition technology, the voice recognition information includes a voice time period and a non-voice time period, and the abnormal data packet information includes a total amount of abnormal data packets, a time point of the abnormal data packets, and a total amount of data packets of the target voice call; The state type information of the media address is obtained according to the call quality score, human voice recognition information and abnormal data packet information of the target voice call, including: According to the human voice time period, the non-human voice time period, the time point of the abnormal data packet and the total number of abnormal data packets, the proportion of abnormal data packets in the human voice time period is obtained; Obtaining a proportion of abnormal data packets according to the total amount of abnormal data packets and the total amount of data packets; When the call quality score of the target voice call is greater than or equal to the network anomaly score threshold, determining the state type information of the media address as one of a business type fluctuation anomaly, a network type fluctuation anomaly, and a normal state according to the proportion of abnormal data packets in the human voice time period of the target voice call and the proportion of abnormal data packets in the target voice call; The call routing of the media address is controlled according to the status type information.

2. The method according to claim 1, characterized in that The acquiring the status type information of the media address includes: When the call quality score of the target voice call is less than the network anomaly score threshold, the status type information of the media address is set to network anomaly.

3. The method according to claim 1, characterized in that The determining the state type information of the media address according to the proportion of abnormal data packets in the human voice time period in the target voice call and the proportion of abnormal data packets in the target voice call includes: When the proportion of abnormal data packets in the voice time period is greater than or equal to the voice data packet abnormality threshold, and the proportion of abnormal data packets is greater than or equal to the service abnormality threshold, the status type information of the media address is set to service-type fluctuation abnormality.

4. The method according to claim 1, characterized in that: The determining of the state type information of the media address according to the proportion of abnormal data packets in the human voice time period in the target voice call and the proportion of abnormal data packets in the target voice call also includes: When the proportion of abnormal data packets in the human voice time period is less than the human voice data packet abnormality threshold, and the proportion of abnormal data packets is greater than or equal to the network abnormality threshold, the state type information of the media address is set to network type fluctuation abnormality.

5. The method according to claim 1, characterized in that The determining of the state type information of the media address according to the proportion of abnormal data packets in the human voice time period in the target voice call and the proportion of abnormal data packets in the target voice call also includes: When the proportion of abnormal data packets in the voice time period is greater than or equal to the voice data packet abnormality threshold and the proportion of abnormal data packets is less than the service abnormality threshold, the state type information of the media address is set to a normal state.

6. The method according to claim 1, characterized in that The determining of the state type information of the media address according to the proportion of abnormal data packets in the human voice time period in the target voice call and the proportion of abnormal data packets in the target voice call also includes: When the proportion of abnormal data packets in the human voice time period is less than the human voice data packet abnormality threshold and the proportion of abnormal data packets is less than the network abnormality threshold, the state type information of the media address is set to a normal state.

7. The method according to claim 1, characterized in that The controlling the call routing of the media address according to the status type information includes: If the number of voice calls associated with the media address is greater than or equal to the monitoring threshold within the preset time period, the number of occurrences of various status type information of the media address within the preset time period is obtained; The call routing of the media address is controlled according to the number of occurrences of various status type information of the media address.

8. The method according to claim 7, characterized in that The controlling the call routing of the media address according to the number of occurrences of various status type information of the media address includes: If the number of occurrences of any one or more of the various status type information is greater than or equal to the corresponding threshold or the corresponding preset proportion threshold, an alarm message is generated and the routing of the media address is stopped.

9. The method according to claim 8, characterized in that If the number of occurrences of any one or more of the various status type information is greater than or equal to the corresponding threshold or the corresponding preset proportion threshold, an alarm message is generated and routing of the media address is stopped, including: When the proportion of the number of network anomalies within the preset time period is greater than a first preset proportion threshold, an alarm message is generated and the routing of the media address is stopped, wherein the proportion of the number of network anomalies is obtained by the total number of various status type information within the preset time period and the number of network anomalies; or When the proportion of the number of abnormal business-type fluctuations is greater than a second preset proportion threshold, an alarm message is generated and routing of the media address is stopped, wherein the proportion of the number of abnormal business-type fluctuations is obtained by the total number and the number of abnormal business-type fluctuations; or When the proportion of the number of network type fluctuation anomalies is greater than the third preset proportion threshold, an alarm message is generated and the routing of the media address is stopped, wherein the proportion of the number of network type fluctuation anomalies is obtained by the total number and the number of network type fluctuation anomalies.

10. A device for monitoring the voice call quality of a media address, characterized in that: It includes an acquisition module, a state type acquisition module and a control module, wherein: The acquisition module is configured to acquire a call record generated by an outbound call system recording a target voice call, wherein the call record includes a call quality score, voice recognition information, abnormal data packet information, and a media address of the target voice call, wherein the voice recognition information is generated by an automatic voice recognition technology for the target voice call, the voice recognition information includes a voice time period and a non-voice time period, and the abnormal data packet information includes a total amount of abnormal data packets, a time point of abnormal data packets, and a total amount of data packets of the target voice call; The state type acquisition module is configured to acquire the state type information of the media address according to the call quality score, human voice recognition information and abnormal data packet information of the target voice call, and is further configured to acquire the proportion of abnormal data packets in the human voice time period according to the human voice time period, non-human voice time period, abnormal data packet time point and the total amount of abnormal data packets; acquire the proportion of abnormal data packets according to the total amount of abnormal data packets and the total amount of data packets; when the call quality score of the target voice call is greater than or equal to the network abnormality score threshold, determine the state type information of the media address as one of business type fluctuation abnormality, network type fluctuation abnormality and normal state according to the proportion of abnormal data packets in the human voice time period in the target voice call and the proportion of abnormal data packets in the target voice call; The control module is configured to control the call routing of the media address according to the status type information.

11. A computer-readable storage medium having a computer program stored thereon, wherein: When the program is executed by a processor, the method according to any one of claims 1 to 9 is implemented.

12. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 9 is implemented.

Images