Judgment method and device for analyzing SIP (Session Initiation Protocol) call result based on signaling and media

Abstract

The invention relates to a method and a device for judging an SIP (Session Initiation Protocol) call result based on common analysis of signaling and media, belongs to the technical field of communication, and solves the problem of call result judgment. The method comprises the following steps: obtaining SIP signaling and RTP media data; capturing and analyzing the RTP media data, and converting an original media stream into a media file; extracting features from the media file and generating a corresponding character text file; acquiring an SIP signaling message by capturing and analyzing the SIP signaling; dividing the call into stages according to the SIP message; in each stage, the SIP signaling message and the text file information extracted from the RTP media data are combined for analysis and judgment, a calling result is obtained, and calling failure reasons are accurately positioned. According to the invention, the SIP signaling process is analyzed in a segmented manner, and the SIP signaling and the RTP media stream are analyzed in a combined manner, so that the accuracy of call result judgment is improved.

Description

Technical Field

[0001] The present invention belongs to the field of communication technology, and in particular relates to a method and device for determining a SIP call result based on joint analysis of signaling and media. Background Art

[0002] The SIP protocol is a signaling control protocol widely used in voice call modes such as IMS, VoLTE, and VoNR. The signaling monitoring system collects and analyzes the signaling between the interfaces in the above communication network to restore the entire process of calls, registrations, and other services. The signaling monitoring system generates CDR (Call Detail Records) for a call by analyzing the signaling protocol. The record fields include: start and end time, calling number, called number, source network element address, destination network element address, call duration, call result, release reason, etc. CDR is convenient for directly understanding the call process. Among them, the fields except the call result field are directly analyzed and judged by signaling direction, parameters, and time. The call result field is a field calculated and analyzed indirectly.

[0003] The determination of the call result field, especially the determination of the call loss result of the abnormal call process, is the most important field in the CDR. It can intuitively reflect whether a call is successful or failed, and if it fails, what kind of failure it is. The call result field requires a comprehensive judgment of the sequence, direction, signaling reason value, message parameters, etc. of all signaling messages in the entire call process.

[0004] The call results can provide network maintenance personnel with information about network quality and end-user call quality, and can also provide customer service personnel with information about end-user call complaints.

[0005] The existing SIP call result judgment only includes a few simple judgments, including successful call, busy line, unreachable, wrong call, and call timeout. The call result judgment is not rich and can only meet the basic use requirements of network maintenance personnel and customer service personnel, and cannot be more accurate to reflect the call result status type. The existing judgment is completely based on signaling messages. The message flow of a call process is judged, and the error code of the SIP protocol is used for analysis. Although the SIP protocol defines various error codes, sometimes these error codes are not clear enough or have multiple meanings. This may lead to misunderstanding or wrong judgment of the cause of call failure. SIP calls may involve communications between multiple networks and operators, which increases the complexity of cross-network judgment. Different operators have different understandings of the protocol, which will cause different error codes to be returned for an abnormal call, so judging the call result based solely on signaling may lead to inconsistent judgments.

[0006] In summary, although existing technologies and tools can help determine SIP call results, there are still some problems such as incompleteness, only broad call result determination, ambiguity in signaling error codes, inconsistent determination of cross-network problems, etc. Therefore, ensuring accurate determination of call results requires comprehensive consideration of these factors and the adoption of appropriate methods and strategies. Summary of the invention

[0007] In view of the above analysis, the present invention aims to disclose a method and device for determining SIP call results based on joint analysis of signaling and media; by performing segmented analysis on the SIP signaling process and combining analysis of SIP signaling and RTP media streams, the accuracy of call result determination is improved.

[0008] The present invention discloses a method for determining a SIP call result based on joint analysis of signaling and media, comprising:

[0009] Step S1, obtaining SIP signaling and RTP media data;

[0010] Step S2: converting the original media stream into a media file by capturing and analyzing the RTP media data; extracting features from the media file and generating a corresponding text file;

[0011] Step S3, obtaining a SIP signaling message by capturing and parsing the SIP signaling; dividing the call into stages according to the SIP signaling message;

[0012] Step S4: In each stage, the SIP signaling message is combined with the text file information extracted from the RTP media data to perform analysis and judgment, obtain the call result, and accurately locate the cause of the call failure.

[0013] Furthermore, the step S2 includes:

[0014] Step S201: Use a media probe to capture and analyze real-time RTP media data provided by the operator network element; encode the voice data in the media stream including the abnormal notification sound into an audio format including WAV and MP3, and save it as an audio media file;

[0015] Step S202: extract features from the media file to generate a text file containing additional information including punctuation marks and timestamps.

[0016] Furthermore, the step S202 includes:

[0017] 1) Extract the audio signal from the audio media file and convert it into Mel spectrum as input feature;

[0018] 2) using an encoder to process the Mel frequency spectrum of the input representation, perform feature extraction and generate an implicit representation;

[0019] The Mel spectrum is input into the Transformer-based model for encoding to generate the encoded latent representation:

[0020] 3) using an autoregressive decoder to gradually generate a text sequence for the generated implicit representation;

[0021] At each step of autoregressive decoding, the next text token is generated based on the output of the previous decoder and the hidden representation Z of the encoder; and the cross-attention mechanism is used to ensure that the generated text is consistent with the audio content.

[0022] Furthermore, in step S4, the call is divided into three stages using step S3: call start, pre-ringing stage, ringing stage, and answering stage; in each stage, the call result is judged based on the calling user's call perception, whether a ringback tone or a color ringback tone is heard, and whether a conversation has already been established with the called party.

[0023] Furthermore, the determination of the call results in the pre-ringing stage before the call is started includes:

[0024] 1) For call results of non-existent or empty user, absent user, busy user, no service on the called side, business restrictions and incoming call restrictions, the SIP signaling and parameters are used as the signaling judgment basis and the keywords recognized in the text are used as the media judgment basis. The combination of the two or the combination of the two is used as the basis for classifying the call results for judgment and classification;

[0025] 2) For call cancellation / early release of the caller, request timeout and network failure, the call results are judged and classified based on the obtained SIP signaling and parameters as the signaling judgment basis;

[0026] 3) For the call results of suspension and arrears, the keywords recognized in the text are used as the basis for media judgment to make judgments and classifications.

[0027] Furthermore, the determination of each call result included in the ringing stage includes:

[0028] 1) For call rejection and long-term unanswered call results, the SIP signaling and parameters are used as the signaling judgment basis and the keywords recognized in the text are used as the media judgment basis. The two are combined or used alone as the basis for call result classification to make judgment and classification;

[0029] 2) For the call results of call waiting and ringing release, the SIP signaling and parameters are obtained as the signaling judgment basis for judgment and classification.

[0030] Furthermore, the judgment of each call result included in the answering stage includes:

[0031] 1) For call results of call hold and call drop, the SIP signaling and parameters are used as the signaling judgment basis and the keywords identified in the text are used as the media judgment basis. The two are combined or used alone as the basis for call result classification to make judgment and classification;

[0032] 2) For the call result of the ping response, the SIP signaling and parameters are obtained as the signaling judgment basis for judgment and classification.

[0033] The present invention also discloses a determination device for analyzing SIP call results based on signaling and media, comprising:

[0034] Operator network elements, RTP media probes, speech-to-text servers, and signaling probes;

[0035] The operator network element is used to provide SIP signaling and RTP media data;

[0036] The RTP media probe is used to process the media data of the abnormal notification sound and convert the original media stream into a media file;

[0037] The speech-to-text server is used to process and map the RTP audio stream in the media file output by the RTP media probe into a text sequence, and output a generated text file;

[0038] The signaling probe is used to monitor and analyze SIP signaling; capture and parse SIP signaling messages transmitted through the network, and analyze the establishment, maintenance and release processes of calls; divide calls into different stages according to SIP signaling messages; receive text files converted by a voice-to-text server; in different SIP call stages, combine SIP signaling message parameters and text files to analyze and locate call results, obtain call results, and accurately locate the cause of call failure and output it to the user.

[0039] Furthermore, the speech-to-text server comprises: a feature extraction module, an encoder module and a decoder module;

[0040] The feature extraction module is used to extract the audio signal from the audio media file output by the RTP media probe and convert it into a Mel spectrum as an input feature;

[0041] The encoder module is used to process the Mel spectrum of the input representation, encode it based on the Transformer model, and generate an implicit representation after feature extraction;

[0042] The decoder module is used to perform autoregressive decoding on the generated implicit representation to gradually generate a text sequence;

[0043] At each step of autoregressive decoding, the next text token is generated based on the output of the previous decoder and the hidden representation Z of the encoder; and the cross-attention mechanism is used to ensure that the generated text is consistent with the audio content.

[0044] Furthermore, while capturing and parsing SIP signaling, the signaling probe also receives media data captured by the RTP media probe that is converted into a text file by the voice-to-text server; and performs call result analysis and positioning in combination with the SIP signaling message and the text file;

[0045] The SIP call pre- and post-call processes are used to divide a call into three stages: call start, pre-ringing stage, ringing stage, and answering stage. In each stage, the call result is determined based on the calling user's call perception, whether a ringback tone or a ringback tone is heard, and whether a call has been made with the called party.

[0046] The present invention can achieve the following beneficial effects:

[0047] 1) The present invention reasonably combines signaling and media analysis to determine the call result, and can determine the specific call result through media file analysis, which is originally impossible to accurately determine using only signaling. It can give an accurate result conclusion for call failures that were originally impossible to locate, and communication call results can be more accurately and intuitively presented to communication network maintenance practitioners.

[0048] 2) Currently, there are only a few simple call result judgments in the industry. The present invention divides the call into three stages for call result judgment through signaling analysis, avoiding confusion in the judgment results, and subdividing the call results into dozens of types, effectively allowing network maintenance personnel and customer service personnel to obtain more detailed call failure conclusions and positioning, which is beneficial to daily network operation and maintenance, evaluating network call quality, and can better provide network service guarantees for users. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The drawings are only for the purpose of illustrating particular embodiments and are not to be considered limiting of the present invention. Like reference symbols denote like components throughout the drawings.

[0050] Figure 1 A flow chart of a method for determining a SIP call result based on joint analysis of signaling and media in an embodiment of the present invention;

[0051] Figure 2 It is a schematic diagram of the process of SIP call before and after in the embodiment of the present invention;

[0052] Figure 3 The figure is a schematic diagram showing the connection components of a determination device for analyzing a SIP call result based on signaling and media in an embodiment of the present invention. DETAILED DESCRIPTION

[0053] The preferred embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used to illustrate the principles of the present invention together with the embodiments of the present invention.

[0054] Embodiment 1

[0055] An embodiment of the present invention discloses a method for determining a SIP call result based on analyzing both signaling and media. Figure 1 As shown, including:

[0056] Step S1, obtaining SIP signaling and RTP media data;

[0057] Step S2: converting the original media stream into a media file by capturing and analyzing the RTP media data; extracting features from the media file and generating a corresponding text file;

[0058] Step S3, obtaining a SIP signaling message by capturing and parsing the SIP signaling; dividing the call into stages according to the SIP signaling message;

[0059] Step S4: In each stage, the SIP signaling message is combined with the text file information extracted from the RTP media data to perform analysis and judgment, obtain the call result, and accurately locate the cause of the call failure.

[0060] Specifically, in step S1, the obtained SIP signaling and RTP media data are provided by the operator network element, and the data provided by the operator network element includes data of voice call modes including fixed network IMS, VoLTE, and VoNR.

[0061] Specifically, step S2 includes:

[0062] Step S201: Use a media probe to capture and analyze real-time RTP media data provided by the operator network element; encode the voice data in the media stream including the abnormal notification sound into an audio format including WAV and MP3, and save it as an audio media file;

[0063] The media probe is mainly used to process the media data of the abnormal notification sound. The media probe can convert the original media stream into a media file. In this process, the media probe encodes the voice data in the media data into an appropriate audio format and saves it as an audio file, such as WAV, MP3 and other formats.

[0064] Step S202: extracting features from the media file to generate a text file containing additional information including punctuation marks and timestamps;

[0065] Specifically include:

[0066] 1) Extract the audio signal from the audio media file and convert it into Mel spectrum as input feature;

[0067] Convert the audio signal S(t) into a Mel-spectrogram as input:

[0068] X = Mel-Spectrogram(S(t));

[0069] Where S(t) is the RTP audio data, and X is the Mel spectrum representation;

[0070] 2) using an encoder to process the Mel frequency spectrum of the input representation, perform feature extraction and generate an implicit representation;

[0071] Input the Mel spectrum X into the Transformer-based model for encoding to generate the encoded latent representation Z:

[0072] The encoding process can be expressed by the following formula:

[0073] Z = Encoder (X);

[0074] 3) using an autoregressive decoder to gradually generate a text sequence for the generated implicit representation;

[0075] In each step of autoregressive decoding, the next text token is generated based on the output of the previous decoder and the hidden representation Z of the encoder; and the cross-attention mechanism is used to ensure that the generated text is consistent with the audio content.

[0076] The decoding process can be expressed by the following formula:

[0077] T(t)=Decoder(Z).

[0078] Specifically, in step S3, the SIP signaling is captured and parsed using a signaling probe to obtain a SIP signaling message;

[0079] The signaling probe is a device for monitoring and analyzing SIP signaling. It can capture and parse SIP signaling messages transmitted through the network so as to analyze the establishment, maintenance and release process of calls.

[0080] In this embodiment, the signaling probe analyzes the media file and the text file together to determine the call result. The signaling probe sends the call result to the upper layer application user, such as network maintenance personnel, customer service personnel, etc.

[0081] Specifically, in step S4, the call is divided into three stages using step S3: call start, pre-ringing stage, ringing stage, and answering stage; in each stage, the call result is judged based on the calling user's call perception, whether a ringback tone or a color ringback tone is heard, and whether a conversation has been established with the called party.

[0082] like Figure 2 As shown, in the three stages of a call,

[0083] Start calling, pre-ringing stage;

[0084] In this phase, the calling user sends a SIP INVITE request to the called user to request to establish a call connection. This phase also includes the exchange of SDP (Session Description Protocol) information, which is used to negotiate the media parameters and codec information of the call.

[0085] Ringing stage: Once the called party accepts the call invitation, the called terminal will return a 180Ring message, indicating that the called terminal has rung, and returns a ringback tone to the calling side or the calling party hears a ringback tone. This means that the call has been connected and the routing path of each node from end to end has been opened.

[0086] Answering stage: After the ringing, the called party picks up the phone or presses the answer button, and the conversation officially begins. During the conversation stage, the calling and called parties will exchange real-time audio and video media data through the RTP (Real-time Transport Protocol) protocol.

[0087] RTP can transmit both the call media data of the calling party and the called party, as well as the notification tone of the call failure. When the RTP notification tone is transmitted to the calling user, it is usually used to notify the user of a new event or remind the user to perform certain operations. These notification tones can be short sound clips, such as "beep" or "beep", or customized voice prompts, such as "the phone you dialed is in a call" or "the phone you dialed is powered off". In general, RTP notification tones play an important role in real-time communication systems, helping users to keep abreast of important information, call status or event reminders, and improving user experience.

[0088] The call results are judged according to the three call stages, which can effectively avoid confusion in judging different types of results, and each call result is clearer and more intuitive.

[0089] Specifically, the determination of each call result in the pre-ringing stage of starting a call includes:

[0090] 1) For call results of non-existent or empty user, absent user, busy user, no service on the called side, business restrictions and incoming call restrictions, the SIP signaling and parameters are used as the signaling judgment basis and the keywords recognized in the text are used as the media judgment basis. The combination of the two or the combination of the two is used as the basis for classifying the call results for judgment and classification;

[0091] 2) For call cancellation / early release of the caller, request timeout and network failure, the call results are judged and classified based on the obtained SIP signaling and parameters as the signaling judgment basis;

[0092] 3) For the call results of suspension and arrears, the keywords recognized in the text are used as the basis for media judgment to make judgments and classifications.

[0093] The corresponding table of call result classification and judgment basis at the beginning of the call and before the ringing is shown in Table 1:

[0094] Table 1

[0095]

[0096]

[0097]

[0098] Specifically, the determination of each call result in the ringing stage includes:

[0099] 1) For call rejection and long-term unanswered call results, the SIP signaling and parameters are used as the signaling judgment basis and the keywords recognized in the text are used as the media judgment basis. The two are combined or used alone as the basis for call result classification to make judgment and classification;

[0100] 2) For the call results of call waiting and ringing release, the SIP signaling and parameters are obtained as the signaling judgment basis for judgment and classification.

[0101] The corresponding table of call result classification and judgment basis in the ringing stage is shown in Table 2:

[0102] Table 2

[0103]

[0104] Specifically, the judgment of each call result included in the answering stage includes:

[0105] 1) For call results of call hold and call drop, the SIP signaling and parameters are used as the signaling judgment basis and the keywords identified in the text are used as the media judgment basis. The two are combined or used alone as the basis for call result classification to make judgment and classification;

[0106] 2) For the call result of the ping response, the SIP signaling and parameters are obtained as the signaling judgment basis for judgment and classification.

[0107] The corresponding table of call result classification and judgment basis in the answering stage is shown in Table 3:

[0108] Table 3

[0109]

[0110] In summary, the embodiments of the present invention reasonably combine signaling and media analysis to determine the call result, and can determine the specific call result through media file analysis, which is originally impossible to accurately determine using only signaling. It allows accurate result conclusions to be given for call failures that were originally impossible to locate, and communication call results can be more accurately and intuitively presented to communication network maintenance practitioners.

[0111] Currently, there are only a few simple call result judgments in the industry. The embodiment of the present invention divides the call into three stages for call result judgment through signaling analysis, thereby avoiding confusion in the judgment results and subdividing the call results into dozens of types, which effectively allows network maintenance personnel and customer service personnel to obtain more detailed call failure conclusions and positioning, which is beneficial to daily network operation and maintenance, evaluates network call quality, and can better provide network service guarantees for users.

[0112] Embodiment 2

[0113] An embodiment of the present invention discloses a determination device based on the joint analysis of SIP call results by signaling and media, such as Figure 3 As shown, it includes: operator network element, RTP media probe, speech-to-text server and signaling probe;

[0114] Operator network elements, RTP media probes, speech-to-text servers, and signaling probes;

[0115] The operator network element is used to provide SIP signaling and RTP media data;

[0116] The RTP media probe is used to process the media data of the abnormal notification sound and convert the original media stream into a media file;

[0117] The speech-to-text server is used to process and map the RTP audio stream in the media file output by the RTP media probe into a text sequence, and output a generated text file;

[0118] The signaling probe is used to monitor and analyze SIP signaling; capture and parse SIP signaling messages transmitted through the network, and analyze the establishment, maintenance and release processes of calls; divide calls into different stages according to SIP signaling messages; receive text files converted by a voice-to-text server; in different SIP call stages, combine SIP signaling message parameters and text files to analyze and locate call results, obtain call results, and accurately locate the cause of call failure and output it to the user.

[0119] Specifically, the data provided by the operator network element includes SIP signaling and RTP media data of voice call modes including fixed network IMS, VoLTE, and VoNR.

[0120] Specifically, the RTP media probe captures and analyzes real-time RTP media data provided by the operator network element; encodes voice data in the media stream including the abnormal notification sound into audio formats including WAV and MP3, and saves it as an audio media file.

[0121] Specifically, the speech-to-text server includes: a feature extraction module, an encoder module and a decoder module;

[0122] The feature extraction module is used to extract the audio signal from the audio media file output by the RTP media probe and convert it into a Mel spectrum as an input feature;

[0123] Convert the audio signal S(t) into a Mel-spectrogram as input:

[0124] X = Mel-Spectrogram(S(t));

[0125] Where S(t) is the RTP audio data, and X is the Mel spectrum representation;

[0126] The encoder module is used to process the Mel spectrum of the input representation, encode it based on the Transformer model, and generate an implicit representation after feature extraction;

[0127] Input the Mel spectrum X into the Transformer-based model for encoding to generate the encoded latent representation Z:

[0128] The encoding process can be expressed by the following formula:

[0129] Z = Encoder (X);

[0130] The decoder module is used to perform autoregressive decoding on the generated implicit representation to gradually generate a text sequence;

[0131] At each step of autoregressive decoding, the next text token is generated based on the output of the previous decoder and the hidden representation Z of the encoder; and the cross-attention mechanism is used to ensure that the generated text is consistent with the audio content.

[0132] The decoding process can be expressed by the following formula:

[0133] T(t)=Decoder(Z).

[0134] Specifically, the signaling probe is a device for monitoring and analyzing SIP signaling, which can capture and parse SIP signaling messages transmitted through the network so as to analyze the establishment, maintenance and release process of a call.

[0135] In this embodiment, the signaling probe, while capturing and parsing SIP signaling, also receives media data captured by the RTP media probe converted into a text file by the voice-to-text server; and performs call result analysis and positioning in combination with the SIP signaling message and the text file;

[0136] The SIP call pre- and post-call processes are used to divide a call into three stages: call start, pre-ringing stage, ringing stage, and answering stage. In each stage, the call result is determined based on the calling user's call perception, whether a ringback tone or a ringback tone is heard, and whether a call has been made with the called party.

[0137] Specifically, when starting a call, the judgment of each call result in the pre-ringing stage includes:

[0138] 1) For call results of non-existent or empty user, absent user, busy user, no service on the called side, business restrictions and incoming call restrictions, the SIP signaling and parameters are used as the signaling judgment basis and the keywords recognized in the text are used as the media judgment basis. The combination of the two or the combination of the two is used as the basis for classifying the call results for judgment and classification;

[0139] 2) For call cancellation / early release of the caller, request timeout and network failure, the call results are judged and classified based on the obtained SIP signaling and parameters as the signaling judgment basis;

[0140] 3) For the call results of suspension and arrears, the keywords recognized in the text are used as the basis for media judgment to make judgments and classifications.

[0141] The judgment of each call result in the ringing stage includes:

[0142] 1) For call rejection and long-term unanswered call results, the SIP signaling and parameters are used as the signaling judgment basis and the keywords recognized in the text are used as the media judgment basis. The two are combined or used alone as the basis for call result classification to make judgment and classification;

[0143] 2) For the call results of call waiting and ringing release, the SIP signaling and parameters are obtained as the signaling judgment basis for judgment and classification.

[0144] The judgment of each call result in the answering stage includes:

[0145] 1) For call results of call hold and call drop, the SIP signaling and parameters are used as the signaling judgment basis and the keywords identified in the text are used as the media judgment basis. The two are combined or used alone as the basis for call result classification to make judgment and classification;

[0146] 2) For the call result of the ping response, the SIP signaling and parameters are obtained as the signaling judgment basis for judgment and classification.

[0147] In this embodiment, please refer to Embodiment 1 for more specific technical details.

[0148] In summary, the embodiments of the present invention reasonably combine signaling and media analysis to determine the call result, and can determine the specific call result through media file analysis, which is originally impossible to accurately determine using only signaling. It allows accurate result conclusions to be given for call failures that were originally impossible to locate, and communication call results can be more accurately and intuitively presented to communication network maintenance practitioners.

[0149] Currently, there are only a few simple call result judgments in the industry. The embodiment of the present invention divides the call into three stages for call result judgment through signaling analysis, thereby avoiding confusion in the judgment results and subdividing the call results into dozens of types, which effectively allows network maintenance personnel and customer service personnel to obtain more detailed call failure conclusions and positioning, which is beneficial to daily network operation and maintenance, evaluates network call quality, and can better provide network service guarantees for users.

[0150] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for determining SIP call results based on joint analysis of signaling and media, characterized in that: include: Step S1, obtaining SIP signaling and RTP media data; Step S2: converting the original media stream into a media file by capturing and analyzing the RTP media data; Extract features from media files and generate corresponding text files; Step S3, obtaining a SIP signaling message by capturing and parsing the SIP signaling; Divide the call into stages according to SIP signaling messages; Step S4: In each stage, the SIP signaling message is combined with the text file information extracted from the RTP media data to perform analysis and judgment, obtain the call result, and accurately locate the cause of the call failure.

2. The method for determining the result of a SIP call based on joint analysis of signaling and media according to claim 1, characterized in that: The step S2 includes: Step S201: Use a media probe to capture and analyze real-time RTP media data provided by the operator network element; encode the voice data in the media stream including the abnormal notification sound into an audio format including WAV and MP3, and save it as an audio media file; Step S202: extract features from the media file to generate a text file containing additional information including punctuation marks and timestamps.

3. The method for determining the result of a SIP call based on joint analysis of signaling and media according to claim 2, characterized in that: The step S202 includes: 1) Extract the audio signal from the audio media file and convert it into Mel spectrum as input feature; 2) using an encoder to process the Mel frequency spectrum of the input representation, perform feature extraction and generate an implicit representation; The Mel spectrum is input into the Transformer-based model for encoding to generate the encoded latent representation: 3) using an autoregressive decoder to gradually generate a text sequence for the generated implicit representation; At each step of autoregressive decoding, the next text token is generated based on the output of the previous decoder and the hidden representation Z of the encoder; and the cross-attention mechanism is used to ensure that the generated text is consistent with the audio content.

4. The method for determining the result of a SIP call based on joint analysis of signaling and media according to claim 3, characterized in that: In step S4, the call is divided into three stages using step S3: call start, pre-ringing stage, ringing stage, and answering stage; in each stage, the call result is judged based on the calling user's call perception, whether a ringback tone or a color ringback tone is heard, and whether a conversation has already been established with the called party.

5. The method for determining the result of a SIP call based on joint analysis of signaling and media according to claim 4, characterized in that: The judgment of each call result in the pre-ringing stage includes: 1) For call results of non-existent or empty user, absent user, busy user, no service on the called side, business restrictions and incoming call restrictions, the SIP signaling and parameters are used as the signaling judgment basis and the keywords recognized in the text are used as the media judgment basis. The combination of the two or the combination of the two is used as the basis for classifying the call results for judgment and classification; 2) For call cancellation / early release of the caller, request timeout and network failure, the call results are judged and classified based on the obtained SIP signaling and parameters as the signaling judgment basis; 3) For the call results of suspension and arrears, the keywords recognized in the text are used as the basis for media judgment to make judgments and classifications.

6. The method for determining the result of a SIP call based on joint analysis of signaling and media according to claim 4, characterized in that: The judgment of each call result in the ringing stage includes: 1) For call rejection and long-term unanswered call results, the SIP signaling and parameters are used as the signaling judgment basis and the keywords recognized in the text are used as the media judgment basis. The two are combined or used alone as the basis for call result classification to make judgment and classification; 2) For the call results of call waiting and ringing release, the SIP signaling and parameters are obtained as the signaling judgment basis for judgment and classification.

7. The method for determining the result of a SIP call based on the joint analysis of signaling and media according to claim 4, characterized in that: The answering phase includes the following aspects of the call result determination: 1) For call results of call hold and call drop, the SIP signaling and parameters are used as the signaling judgment basis and the keywords recognized in the text are used as the media judgment basis. The two are combined or used alone as the basis for call result classification to make judgment and classification; 2) For the call result of the ping response, the SIP signaling and parameters are obtained as the signaling judgment basis for judgment and classification.

8. A determination device for SIP call results based on signaling and media analysis, characterized in that: include: Operator network elements, RTP media probes, speech-to-text servers, and signaling probes; The operator network element is used to provide SIP signaling and RTP media data; The RTP media probe is used to process the media data of the abnormal notification sound and convert the original media stream into a media file; The speech-to-text server is used to process and map the RTP audio stream in the media file output by the RTP media probe into a text sequence, and output a generated text file; The signaling probe is used to monitor and analyze SIP signaling; Capture and parse SIP signaling messages transmitted through the network, analyze the call establishment, maintenance and release processes; divide calls into different stages according to SIP signaling messages; receive text files converted by voice-to-text servers; in different SIP call stages, combine SIP signaling message parameters and text files to analyze and locate call results, obtain call results, and accurately locate the cause of call failure and output it to the user.

9. The determination device for SIP call result based on signaling and media analysis according to claim 8, characterized in that: The speech-to-text server comprises: a feature extraction module, an encoder module and a decoder module; The feature extraction module is used to extract the audio signal from the audio media file output by the RTP media probe and convert it into a Mel spectrum as an input feature; The encoder module is used to process the Mel spectrum of the input representation, encode it based on the Transformer model, and generate an implicit representation after feature extraction; The decoder module is used to perform autoregressive decoding on the generated implicit representation to gradually generate a text sequence; At each step of autoregressive decoding, the next text token is generated based on the output of the previous decoder and the hidden representation Z of the encoder; and the cross-attention mechanism is used to ensure that the generated text is consistent with the audio content.

10. The determination device for SIP call results based on signaling and media analysis according to claim 8, characterized in that: While capturing and parsing SIP signaling, the signaling probe also receives media data captured by the RTP media probe that is converted into a text file by the voice-to-text server; and performs call result analysis and positioning in combination with the SIP signaling message and the text file; Using the SIP call pre- and post-call process, a call is divided into three stages: call start, pre-ringing stage, ringing stage, and answering stage; At each stage, the call result is determined based on the calling user's call perception, whether a ringback tone or a color ring tone is heard, and whether a call has been made with the called party.

Images