Message reading method and device, electronic equipment and computer readable storage medium

Abstract

The embodiment of the invention discloses a message reading method and device, electronic equipment and a computer readable storage medium. According to the embodiment of the invention, after at least one to-be-played group message corresponding to a target object in a target group is obtained, a playing sequence and a message type of the to-be-played group message are identified in the target group, then voice conversion is carried out on the to-be-played group message according to the message type, and a monaural target voice message is obtained. Determining a virtual interaction position corresponding to each interaction object in preset virtual interaction positions, then converting the target voice message into a virtual stereo message based on the virtual interaction positions, and playing the virtual stereo message according to a playing sequence; according to the scheme, the reading efficiency of message reading can be improved, the sense of immediacy and interestingness of message reading are enhanced, and the interaction efficiency of message interaction in the group is improved.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a message reading method, device, electronic device and computer-readable storage medium. Background Art

[0002] In recent years, with the rapid development of Internet technology, it has become increasingly convenient for multiple objects to interact through groups. Each object in a group can send interactive messages and use the interactive messages as group messages in the group, thereby realizing multi-party interaction. In multi-party interaction, each object needs to read the group messages in the group. The current message reading method often plays the voice message or text message in the selected group message through mono voice.

[0003] In the process of research and practice of the prior art, the inventor of the present application found that the current message reading method can often only play voice messages or text messages in group messages, while other types of group messages are often unable to play voice. Moreover, the use of mono voice playback cannot accurately distinguish the senders (interaction objects) of different group messages. Under the condition of limited vision, it is impossible to accurately complete multi-party interaction through the read messages, thus resulting in low reading efficiency of message reading. Summary of the invention

[0004] Embodiments of the present invention provide a message reading method, device, electronic device and computer-readable storage medium, which can improve the reading efficiency of message reading.

[0005] A message reading method, comprising:

[0006] Acquire at least one to-be-played group message corresponding to the target object, wherein the to-be-played group message includes an interaction message between at least one interaction object in the target group and the target object;

[0007] Identifying the play order and message type of the to-be-played group message in the target group;

[0008] Performing voice conversion on the to-be-played group message according to the message type to obtain a monophonic target voice message;

[0009] Determining a virtual interaction position corresponding to each interaction object in the preset virtual interaction position, wherein the virtual interaction position indicates a sound playing direction of an interaction message corresponding to the interaction object;

[0010] Based on the virtual interaction position, the target voice message is converted into a virtual stereo message, and the virtual stereo message is played according to the playing order.

[0011] Accordingly, an embodiment of the present invention provides a message reading device, including:

[0012] An acquisition unit, configured to acquire at least one to-be-played group message corresponding to a target object, wherein the to-be-played group message includes an interaction message between at least one interaction object in a target group and the target object;

[0013] an identification unit, configured to identify the play order and message type of the to-be-played group message in the target group;

[0014] A conversion unit, configured to perform voice conversion on the to-be-played group message according to the message type to obtain a monophonic target voice message;

[0015] A determination unit, configured to determine a virtual interaction position corresponding to each interaction object in a preset virtual interaction position, wherein the virtual interaction position indicates a sound playing direction of an interaction message corresponding to the interaction object;

[0016] The playing unit is used to convert the target voice message into a virtual stereo message based on the virtual interaction position, and play the virtual stereo message according to the playing order.

[0017] In some embodiments, the determination unit can be specifically used to count the number of interaction messages of each interactive object in the target group, and the number of interaction messages indicates the activity level of the interactive object in the target group; based on the number of interaction messages, determine the virtual interaction position corresponding to each interactive object in the preset virtual interaction position.

[0018] In some embodiments, the determination unit may be specifically configured to filter out target group messages to be played corresponding to each interactive object from the group messages to be played; and count the number of target group messages to be played from the group messages to be played to obtain the number of interactive messages of the interactive object.

[0019] In some embodiments, the determination unit can be specifically used to obtain at least one historical group message within a preset historical time range in the target group; filter out the target historical group messages corresponding to each interactive object from the historical group messages; and count the number of the target historical group messages in the historical group messages to obtain the number of interactive messages for each interactive object.

[0020] In some embodiments, the determining unit may be specifically configured to sort the interaction objects according to the number of interaction messages; and based on the sorting result, filter out a virtual interaction position corresponding to each interaction object from the preset interaction positions.

[0021] In some embodiments, the determining unit may be specifically configured to obtain interaction position configuration information of the target object in the target group; and based on the interaction position configuration information, filter out a virtual interaction position corresponding to each interaction object in the preset virtual interaction positions.

[0022] In some embodiments, the determination unit can be specifically used to display an interaction position configuration page, wherein the interaction position configuration page includes at least one candidate interaction position area corresponding to a preset virtual interaction position and a configuration control corresponding to the interaction object; in response to a move operation on the configuration control, a target interaction position area corresponding to each interaction object is determined in the candidate interaction position area; based on the target interaction position area, the interaction position configuration information of the target object in the target group is generated.

[0023] In some embodiments, the playback unit can be specifically used to obtain a set of azimuth channel data corresponding to at least one preset sound playback orientation; filter out target azimuth channel data corresponding to the virtual interaction position from the azimuth channel data set; and fuse the target azimuth channel data with the target voice message to obtain a virtual stereo message.

[0024] In some embodiments, the playback unit can be specifically used to query the target sound playback direction corresponding to the virtual interaction position in the preset sound playback direction; when the target sound playback direction exists, the azimuth channel data corresponding to the target sound playback direction is filtered out in the azimuth channel data set to obtain the target azimuth channel data; when the target sound playback direction does not exist, the adjacent sound playback direction of the target sound playback direction is determined in the preset sound playback direction, and the azimuth channel data corresponding to the adjacent sound playback direction is extracted in the azimuth channel data set to obtain the target azimuth channel data.

[0025] In some embodiments, the conversion unit can be specifically used to, when the message type of the group message to be played is a voice message, use the group message to be played as a mono target voice message; when the message type of the group message to be played is other types except the voice message, perform voice conversion on the group message to be played to obtain a mono target voice message.

[0026] In some embodiments, the conversion unit can be specifically used to determine the message conversion text of the group message to be played according to the message type; construct the mono voice corresponding to the group message to be played based on the message conversion text to obtain the target voice message.

[0027] In some embodiments, the conversion unit can be specifically used to extract text content from the group message to be played to obtain message conversion text when the message type is a text message; identify message conversion text in the group message to be played when the message type is an image message; and identify interaction scenarios in the target group when the message type is an emoticon message, and based on the interaction scenarios, filter out text corresponding to the group message to be played in a preset text set to obtain message conversion text.

[0028] In some embodiments, the conversion unit can be specifically used to extract acoustic features from the message conversion text based on the object identifier of the interactive object using a speech conversion model; convert the acoustic features into monophonic speech to obtain a target voice message corresponding to the group message to be played.

[0029] In some embodiments, the capture unit can be specifically used to obtain at least one historical voice message sample of the interactive object; identify the text content in the historical voice message sample to obtain the message conversion text sample corresponding to the historical voice message sample; train the preset voice conversion model through the historical voice message sample and the corresponding message conversion text sample to obtain the trained voice conversion model.

[0030] In some embodiments, the conversion unit can be specifically used to filter out a target voice conversion model corresponding to the interactive object from at least one voice conversion model corresponding to a preset sound effect based on the object identifier of the interactive object; and use the target voice conversion model to convert the message conversion text into monophonic voice to obtain a target voice message corresponding to the interactive object.

[0031] In some embodiments, the acquisition unit can be specifically used to display a message interaction page of the target object in the target group, the message interaction page including a list of group messages to be played received within a preset time range; in response to a selection operation on the list of group messages to be played, determining a starting group message to be played corresponding to the selection operation; based on the starting group message to be played, filtering out at least one group message to be played corresponding to the target object in the list of group messages to be played.

[0032] In addition, an embodiment of the present invention further provides an electronic device, including a processor and a memory, wherein the memory stores an application program, and the processor is used to run the application program in the memory to execute the message reading method provided by the embodiment of the present invention.

[0033] In addition, an embodiment of the present invention further provides a computer-readable storage medium, which stores a plurality of instructions, and the instructions are suitable for a processor to load to execute the steps in any message reading method provided by the embodiment of the present invention.

[0034] In addition, an embodiment of the present application also provides a computer program product, including a computer program or instructions, which, when executed by a processor, implements the steps in the message reading method provided in the embodiment of the present application.

[0035] After obtaining at least one to-be-played group message corresponding to a target object in a target group, an embodiment of the present invention identifies the play order and message type of the to-be-played group message in the target group, then performs voice conversion on the to-be-played group message according to the message type to obtain a monophonic target voice message, determines a virtual interaction position corresponding to each interactive object in a preset virtual interaction position, then converts the target voice message into a virtual stereo message based on the virtual interaction position, and plays the virtual stereo message according to the play order; since the scheme can perform voice conversion on the to-be-played group message according to the message type, it can realize the conversion of group messages of various message types into voice messages, and can also determine the virtual interaction position of the interactive object interacting with the target object, so as to convert the converted monophonic target voice message into a virtual stereo message with orientation, and then distinguish different message senders (interaction objects) through the virtual stereo message, so that the message interaction in the target group can be closer to the real communication scene, and therefore, the reading efficiency of message reading can be improved, the sense of presence and fun of message reading can be enhanced, and the interaction efficiency of message interaction in the group can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0037] Figure 1 is a scenario schematic diagram of a message reading method provided by an embodiment of the present invention;

[0038] Figure 2 is a flow chart of a message reading method provided by an embodiment of the present invention;

[0039] Figure 3 is a schematic diagram of training and applying a preset speech conversion model provided by an embodiment of the present invention;

[0040] Figure 4is a histogram of the number of message interactions of different interaction objects provided by an embodiment of the present invention;

[0041] Figure 5 is a page schematic diagram of an interactive position configuration page provided by an embodiment of the present invention;

[0042] Figure 6 is an interaction schematic diagram for determining an interaction position area configured for an interaction object provided by an embodiment of the present invention;

[0043] Figure 7 is a schematic diagram of constructing a virtual stereo message provided by an embodiment of the present invention;

[0044] Figure 8 It is a schematic diagram of the overall process of reading a group message to be played provided by an embodiment of the present invention;

[0045] Fig. 9 is another flow chart of the message reading method provided by an embodiment of the present invention;

[0046] Fig.10 is a structural schematic diagram of a message reading device provided by an embodiment of the present invention;

[0047] Fig.11 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0049] The embodiment of the present invention provides a message reading method, device, electronic device and computer readable storage medium. The message reading device can be integrated in an electronic device, and the electronic device can be a server or a terminal.

[0050] Among them, the server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, network acceleration services (Content Delivery Network, CDN), and basic cloud computing services such as big data and artificial intelligence platforms. The terminal can be a smart phone, tablet computer, laptop computer, desktop computer, smart speaker, smart watch, etc., but is not limited to this. The terminal and the server can be directly or indirectly connected via wired or wireless communication, and this application does not limit this.

[0051] For example, see Figure 1 Taking the message reading device integrated in the electronic device as an example, the electronic device can obtain at least one to-be-played group message corresponding to the target object in the target group, identify the playback order and message type of the to-be-played group message in the target group, and then perform voice conversion on the to-be-played group message according to the message type to obtain a monophonic target voice message, determine the virtual interaction position corresponding to each interaction object in the preset virtual interaction position, and then, based on the virtual interaction position, convert the target voice message into a virtual stereo message, and play the virtual stereo message according to the playback order, thereby improving the reading efficiency of the message reading, enhancing the presence and fun of the message reading, and improving the interaction efficiency of the message interaction in the group.

[0052] It can be understood that in the specific implementation of the present application, relevant data such as the group message to be played of the object, other information of the group message to be played, or information in the target group, when the following embodiments of the present application are applied to specific products or technologies, need to obtain permission or consent, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0053] It should be noted that the description order of the following embodiments is not intended to limit the preferred order of the embodiments.

[0054] This embodiment will be described from the perspective of a message reading device, which can be specifically integrated into an electronic device, which can be a server or a terminal. The terminal can include a tablet computer, a laptop computer, a personal computer (PC), a wearable device, a virtual reality device, or other smart devices that can read messages.

[0055] A message reading method, comprising:

[0056] At least one to-be-played group message corresponding to the target object is obtained, the to-be-played group message includes an interactive message between at least one interactive object in the target group and the target object, the playback order and message type of the to-be-played group messages are identified in the target group, voice conversion is performed on the to-be-played group messages according to the message type to obtain a mono target voice message, a virtual interactive position corresponding to each interactive object is determined in a preset virtual interactive position, the virtual interactive position indicates a sound playback direction of the interactive message corresponding to the interactive object, based on the virtual interactive position, the target voice message is converted into a virtual stereo message, and the virtual stereo message is played according to the playback order.

[0057] like Figure 2 As shown, the specific process of the message reading method is as follows:

[0058] 101. Obtain at least one to-be-played group message corresponding to a target object.

[0059] Among them, the group message to be played includes an interactive message between at least one interactive object in the target group and the target object. The interactive message here has not been triggered by the target object to be played and is in a state to be played. The target group may include the target object and at least one interactive object. The interactive message can be understood as a message sent by the interactive object, which can be an instant message or a non-instant message. There can be multiple types of interactive messages, for example, voice, text, emoticon or image, etc. Therefore, the message type of the group message to be played can also be multiple, for example, voice message, text message, emoticon message or image message, etc.

[0060] There are multiple ways to obtain at least one to-be-played group message corresponding to the target object, which can be specifically as follows:

[0061] For example, at least one group message to be played can be filtered out from the group messages received by the target object in the target group, or at least one group message to be played uploaded by the terminal corresponding to the target object can be obtained, or an instant group message sent by at least one interactive object in the target group that interacts with the target object can be received, and the instant group message can be used as the group message to be played, and so on.

[0062] Among them, there may be multiple ways to filter out at least one group message to be played from the group messages received by the target object in the target group. For example, a message interaction page of the target object in the target group may be displayed, and the message interaction page may include a list of group messages to be played received within a preset time range. In response to a selection operation on the list of group messages to be played, a starting group message to be played corresponding to the selection operation is determined, and based on the starting group message to be played, at least one group message to be played corresponding to the target object is filtered out in the list of group messages to be played.

[0063] Among them, the response is used to indicate the conditions or states on which the executed operations depend. When the dependent conditions or states are met, one or more operations executed can be in real time or have a set delay. Unless otherwise specified, there is no restriction on the order of execution of the multiple operations executed.

[0064] The starting group message to be played may be the group message to be played that starts voice playback, that is, the group messages to be played after the starting group message to be played are all read in the voice playback mode. Based on the starting group message to be played, there may be multiple ways to filter out at least one group message to be played corresponding to the target object in the list of group messages to be played. For example, the group messages to be played after the starting group message to be played may be filtered out in the list of group messages to be played to obtain candidate group messages to be played, and the starting group message to be played and the candidate group messages to be played are used as at least one group message to be played corresponding to the target object.

[0065] 102. Identify the play order and message type of the group messages to be played in the target group.

[0066] The play order may be understood as the order in which the to-be-played group messages are played. The play order may also correspond to the reception time of the to-be-played group messages. The earlier the reception time, the earlier the corresponding play order.

[0067] There are many ways to identify the play order and message type of the group message to be played in the target group, which can be as follows:

[0068] For example, the message attribute information of the group message to be played is obtained, the message format and the receiving time of the group message are identified in the message attribute information, the message type of the group message to be played is determined based on the message format, and the group messages to be played are sorted according to the receiving time, and the playing order of each group message to be played is determined based on the sorting result. Alternatively, the receiving record of the group message to be played can be obtained, and the playing order and message type of the group message to be played can be identified in the receiving record, and so on.

[0069] 103. Perform voice conversion on the group message to be played according to the message type to obtain a mono target voice message.

[0070] Among them, mono refers to an audio format in which the audio signal contains only one channel or channels. In a mono video, all audio information is mixed in one channel and played, and the location and direction of the sound source cannot be distinguished.

[0071] The target voice message may be a group message whose message type is a voice message.

[0072] There are multiple ways to convert the playback group message into voice according to the message type, which can be as follows:

[0073] For example, when the message type of the group message to be played is a voice message, the group message to be played is used as a mono target voice message. When the message type of the group message to be played is other types except voice messages, the group message to be played is voice-converted to obtain a mono target voice message.

[0074] Among them, when the message type of the group message to be played is other types other than voice messages, there can be multiple ways to perform voice conversion on the group message to be played. For example, according to the message type, the message conversion text of the group message to be played is determined, and the mono voice corresponding to the group message to be played is constructed based on the message conversion text to obtain the target voice message.

[0075] Among them, the message conversion text can be understood as the text content corresponding to the target voice message. According to the message type, there are many ways to determine the message conversion text of the group message to be played. For example, when the message type is a text message, the text content is identified in the group message to be played to obtain the message conversion text. When the message type is an image message, the preset image text is used as the message conversion text. When the message type is an emoticon message, the interaction scene is identified in the target group, and based on the interaction scene, the text corresponding to the group message to be played is screened out from the preset text set to obtain the message conversion text.

[0076] Among them, when the message type is a text message, at this time, the text content can be extracted from the group message to be played, and the extracted text content is used as the message conversion text.

[0077] Among them, when the message type is an image message, the message conversion text can be identified in the group message to be played. There are many ways to identify the message conversion text in the group message to be played, for example, a preset image text can be used as the message conversion text of the group message to be played, or an image category can be identified in the group message to be played, and the image text corresponding to the image category can be screened out from the preset image text set to obtain the message conversion text, or the image content can be identified in the group message to be played, and the message conversion text can be determined based on the image content, and so on.

[0078] The preset image text may be a preset image text indicating an image message, for example, "I sent you a picture, please have a look", or other image text indicating that the interactive object sends an image message, and so on.

[0079] The image category may indicate the type of image content in the group message to be played, such as funny, educational, movie, sports or other categories. Taking the image category of sports as an example, there may be multiple image texts selected from the preset image texts, such as "I sent you a sports image, please take a look", or other texts that may indicate the images related to sports sent by the interactive object, etc.

[0080] The image content may be the content of the to-be-played group message indicating that the message type is an image message. There are multiple ways to determine the message conversion text based on the image content, for example, the image content may be directly converted into text content to obtain the message conversion text, or the image text corresponding to the image content may be selected from a preset image text set to obtain the message conversion text, and so on.

[0081] Among them, when the message type is an expression message, the interaction scene can be identified in the target group. The so-called expression message can be a message that represents the expression of the interactive object, for example, it can include messages describing happiness, anger, smiling, laughing, crying or other expressions. The interaction scene can be understood as the interaction context or other scenes in the target group. There are many ways to identify the interaction scene in the target group. For example, you can obtain the historical group message corresponding to the group message to be played, extract the interaction features from the historical group message, and determine the interaction scene based on the interaction features. Alternatively, you can also filter out the context message of the group message to be played whose message type is an expression message in the group message to be played, and determine the interaction scene based on the context message.

[0082] After determining the interaction scenario, the text corresponding to the group message to be played can be filtered out from the preset text set based on the interaction scenario to obtain the message conversion text. There can be many ways to filter out the text corresponding to the group message to be played from the preset text set. For example, you can filter out at least one candidate text corresponding to the interaction scenario from the preset text set to obtain a candidate text set, obtain the expression type of the group message to be played, identify the text corresponding to the expression type in the candidate text set, and obtain the message conversion text. For example, taking the expression of a smiley face as an example, you can filter out the corresponding text such as "hehe" or "smile" from the candidate text set to obtain the message conversion text.

[0083] After determining the message conversion text of the group message to be played, the mono voice corresponding to the group message to be played can be constructed based on the message conversion text to obtain the target voice message. There are many ways to construct the mono voice corresponding to the group message to be played. For example, based on the object identifier of the interactive object, a voice conversion model can be used to extract acoustic features from the message conversion text, and the acoustic features can be converted into a mono voice message to obtain the target voice message corresponding to the group message to be played.

[0084] Among them, the speech conversion model may include a prosody prediction sub-model and an acoustic prediction sub-model. The prosody prediction sub-model is a natural language processing model used to predict the prosody or intonation in the message conversion text. The acoustic prediction sub-model is a model for speech signal processing, which is mainly used to predict the acoustic features in the speech signal, such as the spectrum, sound intensity or fundamental frequency of the speech. Based on the object identification of the interactive object, there are many ways to extract acoustic features from the message conversion text using the speech conversion model. For example, based on the object identification of the interactive object, the prosody prediction sub-model can be used to predict the prosody or intonation information of the message conversion text, thereby obtaining the intonation feature, and the acoustic prediction sub-model can be used to predict the acoustic information of the intonation feature, thereby obtaining the acoustic feature corresponding to the message conversion text.

[0085] After the acoustic features of the message conversion text are extracted using the voice conversion model, the acoustic features can be converted into monophonic speech, thereby obtaining a target voice message corresponding to the group message to be played. There are many ways to convert the acoustic features into monophonic speech. For example, a vocoder can be used to convert the acoustic features into monophonic speech, thereby obtaining a target voice message corresponding to the interactive object.

[0086] In some embodiments, before extracting acoustic features from the message conversion text using a voice conversion model based on the object identification of the interactive object, the preset voice conversion model may be trained to obtain a trained voice conversion model. There may be multiple ways to train the preset voice conversion model, for example, at least one historical voice message sample of the interactive object may be obtained, the text content may be identified in the historical voice message sample, a message conversion text sample corresponding to the historical voice message sample may be obtained, and the preset voice conversion model may be trained using the historical voice message sample and the corresponding message conversion text sample to obtain a trained message conversion model.

[0087] The preset speech conversion model may include a preset prosody prediction submodel and a preset acoustic prediction submodel. There are multiple ways to train the preset speech conversion model through historical speech message samples and corresponding message conversion text samples. For example, the preset prosody prediction submodel of the preset speech conversion model may be used to predict the prosody or intonation information of the message conversion text sample to obtain the sample intonation feature. The preset acoustic prediction submodel of the preset speech conversion model may be used to predict the acoustic information of the sample intonation feature, thereby obtaining the sample acoustic feature corresponding to the message conversion text sample. The sample acoustic feature is converted into monophonic speech using a vocoder to obtain a predicted speech message. The predicted speech message is compared with the historical speech message sample to obtain the speech conversion loss. Based on the speech conversion loss, the preset speech conversion model is converged to obtain the trained speech conversion model.

[0088] The speech conversion loss can be understood as the loss generated when performing speech conversion on the message conversion text sample. There are many ways to compare the predicted speech message with the historical speech message sample to obtain the speech conversion loss. For example, speech features can be extracted from the historical speech message sample to obtain the historical speech features, and speech features can be extracted from the predicted speech message to obtain the predicted speech features, and the feature similarity or feature distance between the predicted speech features and the historical speech features can be calculated, and the speech conversion loss can be determined based on the feature similarity or feature distance, etc.

[0089] The training corpus is constructed through the historical voice message samples of the interaction object and the corresponding message conversion text samples, and the preset voice conversion model is trained through the prediction library, so that the trained voice conversion model can construct a personalized voice message that matches the interaction object, that is, the constructed voice message matches the timbre of the interaction object. The process of training and applying the preset voice conversion model can be as follows: Figure 3 As shown, in the training phase, by constructing a corpus of each interactive object, the preset prosody prediction submodel and the preset acoustic prediction submodel in the preset speech conversion model are trained, thereby obtaining a trained prosody prediction submodel and a trained acoustic prediction submodel, and the trained prosody prediction submodel and the trained acoustic prediction submodel are used as the trained speech conversion model. In the application process of the speech conversion model, the trained prosody prediction submodel is used to predict the prosody information of the message conversion text (text to be processed) to obtain the intonation feature, the trained acoustic prediction submodel is used to predict the acoustic information of the intonation feature to obtain the acoustic feature, and the voice coder is used to generate a speech message corresponding to the acoustic feature to obtain the target speech message corresponding to the group message to be played.

[0090] In some embodiments, based on the object identifier of the interactive object, a target voice conversion model corresponding to the interactive object can be screened out from the voice conversion model corresponding to at least one preset sound effect, and the target voice conversion model can be used to convert the message conversion text into monophonic voice to obtain the target voice message corresponding to the interactive object.

[0091] Among them, the preset sound effect can be a pre-set sound effect, and there can be multiple types of sound effects, for example, it can include loli voice, middle-aged male voice, middle-aged female voice, child voice or other sound effects, etc. The voice conversion model corresponding to each preset sound effect can be understood as a target voice message that can convert the message conversion text into a specific sound effect (i.e., the preset sound effect). Based on the object identification of the interactive object, there can be multiple ways to filter out the target voice conversion model corresponding to the interactive object from the voice conversion model corresponding to at least one preset sound effect. For example, based on the object identification of the interactive object, the object type of the interactive object can be determined, and the target sound effect corresponding to the object type can be filtered out from the preset sound effects, and the voice conversion model corresponding to the target sound effect is used as the target voice conversion model corresponding to the interactive object.

[0092] Among them, the object type can correspond to the preset sound effect. For example, taking the preset sound effect as a middle-aged male voice as an example, the corresponding object type is a middle-aged man. Taking the preset sound effect as a child's voice as an example, the corresponding object type can be a child, and so on.

[0093] Among them, using the target speech conversion model to convert the message conversion text into monophonic speech is similar to using the speech conversion model to perform speech conversion, as described above, and will not be repeated here.

[0094] 104. Determine a virtual interaction position corresponding to each interaction object in the preset virtual interaction positions.

[0095] Among them, the virtual interaction position indicates the sound playback direction of the interaction message of the interactive object. Different interactive objects have different virtual interaction positions relative to the target object in the target group. The virtual interaction position here can also be understood as the interaction position of other interactive objects relative to the target object with the target object as the center when interacting with the target object in the target group. This interaction position is a virtual interaction position, which may include interaction direction, interaction distance or position, etc. Since these virtual interaction positions respectively indicate the sound playback direction of the interaction messages sent by different interactive objects, different interactive objects can be distinguished through these virtual interaction positions.

[0096] There are many ways to determine the virtual interaction position corresponding to each interaction object in the preset virtual interaction position, which can be specifically as follows:

[0097] For example, the number of interaction messages of each interaction object in the target group can be counted, and based on the number of interaction messages, the virtual interaction position corresponding to each interaction object can be determined in the preset interaction position. Alternatively, the virtual interaction position corresponding to each interaction object can be randomly determined in the preset virtual interaction position, and so on.

[0098] The number of interactive messages may indicate the activity level of the interactive object in the target group, and may also be understood as the number of interactive messages sent by the interactive object in the target group. There may be multiple ways to count the number of interactive messages of each interactive object in the target group, for example, the target group message to be played corresponding to each interactive object may be screened out from the group messages to be played, the number of the target group messages to be played may be counted in the group messages to be played, and the number of interactive messages of the interactive object may be obtained, or at least one group message within a preset historical time range may be obtained from the target group, the target historical group message corresponding to each interactive object may be screened out from the historical group messages, and the number of interactive messages of each interactive object may be obtained.

[0099] It should be noted that the historical group messages may include historical group messages to be played and historical group messages that have been played.

[0100] After counting the number of interaction messages of each interaction object, the virtual interaction position corresponding to each interaction object can be determined in the preset virtual interaction position based on the number of interaction messages. There are many ways to determine the virtual interaction position corresponding to each interaction object based on the number of interaction messages. For example, the interaction objects can be sorted according to the number of interaction messages, and the virtual interaction position of each interaction object can be screened out in the preset interaction position based on the sorting result.

[0101] There are many ways to sort the interaction objects according to the number of interaction messages. For example, a histogram of the number of interaction messages can be constructed, and the interaction objects can be sorted by the histogram to obtain a sorting result. Alternatively, the interaction objects can be sorted directly from large to small or from small to large based on the number of interaction messages to obtain a sorting result, and so on.

[0102] The histogram of interactive messages can be understood as a histogram showing the number of message interactions of different interactive objects, which can be shown as follows: Figure 4 shown.

[0103] Among them, there can be multiple ways to screen out the virtual interaction position of each interactive object from the preset interaction position based on the sorting result. For example, based on the sorting result, the interactive object with the largest number of message interactions can be assigned the best virtual position (for example, the front or any other preset interaction position set), the interactive object with the second largest number of message interactions can be assigned a suboptimal virtual position (for example, the left front position at an angle of 30 degrees from the front or any other preset interaction position set), the interactive object with the third largest number of message interactions can be assigned a suboptimal virtual position (for example, the right front position at an angle of 30 degrees from the front or any other preset interaction position set), and so on, so that each interactive object is assigned a virtual position, and the virtual position is used as the virtual interaction position; or, based on the sorting result, the interactive object with the smallest number of message interactions can be assigned the best virtual position, the interactive object with the second smallest number of message interactions can be assigned a suboptimal virtual position, and so on, so that each interactive object is assigned a virtual position, and the virtual position is used as the virtual interaction position, and so on.

[0104] The preset virtual interaction position may be a preset interaction orientation, distance or position relative to the target object, etc.

[0105] In some embodiments, the virtual interaction position of the interaction object may be configured to determine the virtual interaction position of each interaction object. For example, the interaction position configuration information of the target object in the target group may be obtained, and based on the interaction position configuration information, the virtual interaction position corresponding to each interaction object may be screened out from the preset virtual interaction positions.

[0106] The interactive position configuration information can be understood as configuration information for configuring the virtual interactive position of each interactive object. There are many ways to obtain the interactive position configuration information of the target object in the target group. For example, an interactive position configuration page can be displayed, and the interactive configuration page includes at least one candidate interactive position area corresponding to the preset virtual interactive position and a configuration control corresponding to the interactive object. In response to a move operation on the configuration control, a target interactive position area corresponding to each interactive object is determined in the candidate interactive position area, and based on the target interactive position area, the interactive position configuration information of the target object in the target group is generated.

[0107] Among them, the response is used to indicate the conditions or states on which the executed operations depend. When the dependent conditions or states are met, one or more operations executed can be in real time or have a set delay. Unless otherwise specified, there is no restriction on the order of execution of the multiple operations executed.

[0108] The candidate interactive location area may be a pre-set interactive location area centered on the target object. The size and position of the interactive location area relative to the target object may be set according to actual application. Figure 5 As shown, it includes at least one candidate interaction position area 51 and a configuration control 52 corresponding to the interaction object. The number of candidate interaction position areas can be greater than or equal to the number of interaction objects, that is, m is less than or equal to n. By moving the configuration control corresponding to the interaction object to the interaction position area, the virtual interaction position corresponding to the interaction position area configured for the interaction object can be determined, and then the interaction position configuration information of the target object in the target group is generated. Based on the interaction position configuration page, the process of determining the interaction position area configured for the interaction object can be as follows: Figure 6 As shown, the target object or other object moves the configuration control of the interactive object to the corresponding candidate position area, and in response to the moving operation of the target object or other object, the candidate position area can be used as the target interactive position area of ​​the interactive object. The virtual interactive position corresponding to the target interactive position area is screened out from the preset virtual interactive position as the virtual interactive position for the interactive object to interact with the target object in the target group. The virtual interactive position of each interactive object or the position information of the virtual interactive position is used as the interactive position configuration information of the target object in the target group. 105. Based on the virtual interactive position, the target voice message is converted into virtual stereo, and the virtual stereo is played according to the playback order.

[0109] Among them, virtual stereo can be understood as simulating the stereo sound signal in reality through technical means. Based on the virtual interaction position, there are many ways to convert the target voice message into virtual stereo, which can be as follows:

[0110] For example, a directional channel data set corresponding to at least one preset sound playback orientation may be obtained, target directional channel data corresponding to a virtual interaction position may be screened out from the directional channel data set, and the target directional channel data may be merged with a target voice message to obtain a virtual stereo message.

[0111] Among them, the channel orientation data set may include at least one azimuth channel data of a preset channel playback orientation. The so-called azimuth channel data can be understood as the data of the corresponding orientation of HRTF (head-related transfer function). The azimuth channel data in the azimuth channel data set is the impact corresponding signal curve of the left and right channels collected according to different wonderful angles. The azimuth channel data set may include HRIR data (a kind of channel data) of the left channel and the right channel. The azimuth channel data set may be an open source data set or a data set collected according to actual applications. Taking the open source data set as an example, there may be multiple types of HRIR open source data, for example, it may include a data set of 125 CIPIC HRTF testers (an open source HRIR data set), a data set of 50 LISTEN HRTF testers (an open source HRIR data set), a MIT KEMAR HRTF data set (an open source HRIR data set), and the like. Taking the channel orientation data set including the MIT KEMAR HRTF data set as an example, the angle of the channel orientation data set is 5 degrees, which means that the HRTF data in the data set is collected from each 5-degree angle, and a total of 72 directions can be included (from 0 degrees to 355 degrees, sampling one direction every 5 degrees). There are many ways to filter out the target orientation channel data corresponding to the virtual interaction position in the orientation channel data, such as, querying the target sound playback position corresponding to the virtual interaction position in the preset channel playback position, when there is a target sound playback position, filtering out the azimuth channel data corresponding to the target sound playback position in the azimuth channel data set, and obtaining the target azimuth channel data, when there is no target sound playback position, determining the adjacent sound playback position of the target sound playback position in the preset sound playback position, and extracting the azimuth channel data corresponding to the adjacent sound playback position in the azimuth channel data, and obtaining the target azimuth channel data.

[0112] Among them, the target sound playback direction can be the sound playback direction corresponding to the virtual interaction position. For example, with the virtual interaction position as the front, at this time, the sound playback direction can be 90 degrees, and so on.

[0113] Among them, when there is a target sound playback direction, the azimuth channel data corresponding to the target sound playback direction can be directly filtered out from the azimuth channel data set to obtain the target azimuth channel data.

[0114] When there is no target sound playing direction, the adjacent sound playing direction of the target sound playing direction can be determined in the preset sound playing direction. There are many ways to determine the adjacent sound playing direction. For example, the direction distance between the target sound direction and each preset sound playing direction can be calculated, and the adjacent sound playing direction can be selected from the preset sound playing directions based on the direction distance.

[0115] Among them, there are various ways to screen adjacent sound playback directions from the preset sound playback directions based on the azimuth distance. For example, the preset sound playback directions can be sorted based on the azimuth distance, and based on the sorting result, the preset sound playback directions with the smallest azimuth distance are screened out from the preset sound playback directions, so as to obtain the adjacent sound playback directions of the target sound playback direction, and so on.

[0116] After screening out the adjacent sound playback directions, the azimuth channel data corresponding to the adjacent sound playback directions can be extracted from the azimuth channel data to obtain the target azimuth channel data.

[0117] After screening out the target azimuth channel data corresponding to the virtual interaction position from the azimuth channel data, the screened target azimuth channel data can be fused with the target voice message to obtain a virtual stereo message. There are various ways to fuse the target azimuth channel data with the target voice message. For example, the target voice message can be convolved with the target azimuth channel data to obtain a two-channel stereo signal, and this stereo signal is used as the virtual stereo message. Specifically, it can be as shown in formula (1), specifically as follows:

[0118]

[0119] Among them, y(n) is the stereo signal (virtual stereo message), u(n) is the monophonic target voice message, and h(n) is the target azimuth channel data corresponding to the virtual interaction position.

[0120] Among them, the process of constructing a virtual stereo message through HRTF can be as Figure 7 shown. The input target voice message is convolved with the target azimuth channel data of the left and right channels respectively, so as to obtain a two-channel stereo signal, and the two-channel stereo signal is used as the virtual stereo message.

[0121] After converting the target voice message into a virtual stereo message based on the virtual interaction position, the virtual stereo message can be played according to the playback order. There are various ways to play the virtual stereo message according to the playback order. For example, the target playback order of the virtual stereo message can be determined based on the playback order, and the virtual stereo message is played according to the target playback order.

[0122] Among them, there are various ways to play the virtual stereo message according to the target playback order. For example, the playback component can be called to play the virtual stereo message according to the target playback order, or the target playback order and the virtual stereo message can be sent to the object terminal corresponding to the target object, so that the object terminal plays the virtual stereo message according to the target playback order, and so on.

[0123] Taking manually selecting at least one group message to be played as an example, the overall process of reading the group message to be played in this solution can be as follows: Figure 8 As shown, the first group message to be played is manually selected from the received group messages, thereby obtaining at least one group message to be played, and the message type of the group message to be played is identified one by one according to the time sequence of the received messages. Since the present scheme will eventually be played in the form of sound, it is necessary to convert the non-needed messages into voice through the TTS method, and the group messages to be played after the voice conversion are uniformly converted into monophonic sound signals (target voice messages). In order to more easily distinguish the message sender (interaction object), the monophonic target voice message can be converted into a virtual stereo message, and then, according to the play order, the virtual stereo message is played. This virtual stereo message with orientation can be closer to the real communication or interaction scene, allowing different message senders (interaction objects) to be set in a fixed orientation as if on the spot, and communicate with the listener (target object or other interaction object), so that the target object can release his hands and eyes, and only needs to listen and record the message through headphones or stereo playback equipment, so that the ordinary group message broadcast is more real, more diverse, and more convenient, thereby improving the reading efficiency of message reading.

[0124] As can be seen from the above, after obtaining at least one group message to be played corresponding to the target object in the target group, the embodiment of the present application identifies the play order and message type of the group message to be played in the target group, and then, according to the message type, performs voice conversion on the group message to be played to obtain a monophonic target voice message, determines the virtual interaction position corresponding to each interactive object in the preset virtual interaction position, and then, based on the virtual interaction position, converts the target voice message into a virtual stereo message, and plays the virtual stereo message according to the play order; because the scheme can perform voice conversion on the group message to be played according to the message type, thereby realizing the conversion of group messages of various message types into voice messages, and further, the virtual interaction position of the interactive object interacting with the target object can be determined, so that the converted monophonic target voice message is converted into a virtual stereo message with orientation, and then different message senders (interaction objects) are distinguished through the virtual stereo message, so that the message interaction in the target group can be closer to the real communication scene, and therefore, the reading efficiency of message reading can be improved, the sense of presence and fun of message reading can be enhanced, and the interaction efficiency of message interaction in the group can be improved.

[0125] According to the method described in the above embodiment, the following is further described in detail with examples.

[0126] In this embodiment, the message reading device is specifically integrated into an electronic device, and the electronic device is a terminal.

[0127] like Fig. 9 As shown, a message reading method, the specific process is as follows:

[0128] 201. A terminal obtains at least one to-be-played group message corresponding to a target object.

[0129] For example, the terminal can display a message interaction page of the target object in the target group, and the message interaction page can include a list of group messages to be played received within a preset time range, and in response to a selection operation on the list of group messages to be played, determine the starting group message to be played corresponding to the selection operation, and filter out the group messages to be played after the starting group message to be played in the list of group messages to be played to obtain candidate group messages to be played, and use the starting group message to be played and the candidate group message to be played as at least one group message to be played corresponding to the target object.

[0130] In some embodiments, the terminal can also obtain at least one group message to be played uploaded by the terminal corresponding to the target object, or can also receive an instant group message sent by at least one interactive object in the target group that interacts with the target object, and use the instant group message as the group message to be played, etc.

[0131] 202. The terminal identifies the play order and message type of the group messages to be played in the target group.

[0132] For example, the terminal can obtain message attribute information of the group message to be played, identify the message format and receiving time of the group message in the message attribute information, determine the message type of the group message to be played based on the message format, and sort the group messages to be played according to the receiving time, and determine the playing order of each group message to be played based on the sorting result. Alternatively, the terminal can also obtain the receiving record of the group message to be played, identify the playing order and message type of the group message to be played in the receiving record, and so on.

[0133] 203. The terminal performs voice conversion on the group message to be played according to the message type to obtain a monophonic target voice message.

[0134] For example, when the message type of the group message to be played is a voice message, the terminal uses the group message to be played as a mono target voice message.

[0135] When the message type is a text message, the terminal identifies the text content in the group message to be played and obtains the message conversion text. When the message type is an image message, the terminal uses the preset image text as the message conversion text of the group message to be played. Alternatively, it can identify the image category in the group message to be played and filter out the image text corresponding to the image category from the preset image text set to obtain the message conversion text. Alternatively, it can also identify the image content in the group message to be played and determine the message conversion text based on the image content, and so on. When the message type is an emoji message, obtain the historical group message corresponding to the group message to be played, extract the interaction features from the historical group message, and determine the interaction scenario based on the interaction features. Alternatively, it can also filter out the context message of the group message to be played whose message type is an emoji message from the group message to be played and determine the interaction scenario based on the context message. Filter out at least one candidate text corresponding to the interaction scenario from the preset text set to obtain a candidate text set, obtain the emoji type of the group message to be played, and identify the text corresponding to the emoji type in the candidate text set to obtain the message conversion text.

[0136] Based on the object identifier of the interaction object, the terminal uses the prosody prediction sub-model of the voice conversion model to predict the prosody or intonation information of the message conversion text, thereby obtaining the intonation feature, and uses the acoustic prediction sub-model of the voice conversion model to predict the acoustic information of the intonation feature, thereby obtaining the acoustic feature corresponding to the message conversion text. The target acoustic feature is converted into a mono voice by a vocoder, thereby obtaining the target voice message corresponding to the interaction object.

[0137] Optionally, in some embodiments, the terminal can also obtain at least one historical voice message sample of the interaction object, identify the text content in the historical voice message sample, and obtain the message conversion text sample corresponding to the historical voice message sample. Use the preset prosody prediction sub-model of the preset voice conversion model to predict the prosody or intonation information of the message conversion text sample to obtain the sample intonation feature, and use the preset acoustic prediction sub-model of the preset voice conversion model to predict the acoustic information of the sample intonation feature, thereby obtaining the sample acoustic feature corresponding to the message conversion text sample. The sample acoustic feature is converted into a mono voice by a vocoder to obtain the predicted voice message. Extract the voice features from the historical voice message sample to obtain the historical voice features, and extract the voice features from the predicted voice message to obtain the predicted voice features. Calculate the feature similarity or feature distance between the predicted voice features and the historical voice features, and determine the voice conversion loss based on the feature similarity or feature distance. Converge the preset voice conversion model based on the voice conversion loss to obtain the trained voice conversion model.

[0138] Optionally, in some embodiments, the terminal can also determine the object type of the interactive object based on the object identifier of the interactive object, filter out the target sound effect corresponding to the object type from the preset sound effects, use the voice conversion model corresponding to the target sound effect as the target voice conversion model corresponding to the interactive object, and use the target voice conversion model to convert the message conversion text into mono voice to obtain the target voice message corresponding to the interactive object.

[0139] 204. The terminal determines the sound playing direction of each interactive object in the preset virtual interactive position.

[0140] For example, the terminal can filter out the target group messages to be played corresponding to each interaction object from the group messages to be played, count the number of target group messages to be played in the group messages to be played, and obtain the number of interaction messages of the interaction object; or, it can obtain at least one group message within a preset historical time range in the target group, filter out the target historical group messages corresponding to each interaction object from the historical group messages, and obtain the number of interaction messages of each interaction object.

[0141] The terminal may construct a histogram of the number of interaction messages, and sort the interaction objects by the histogram to obtain a sorting result, or may directly sort the interaction objects from large to small or from small to large based on the number of interaction messages to obtain a sorting result, and so on.

[0142] Based on the sorting results, the terminal assigns the best virtual position (for example, directly in front or any other preset interaction position) to the interactive object with the largest number of message interactions, the suboptimal virtual position (for example, the left front position at an angle of 30 degrees from the front or any other preset interaction position) to the interactive object with the second largest number of message interactions, the suboptimal virtual position (for example, the right front position at an angle of 30 degrees from the front or any other preset interaction position) to the interactive object with the third largest number of message interactions, and so on, thereby assigning a virtual position to each interactive object and using the virtual position as the virtual interaction position; or, based on the sorting results, the terminal may assign the best virtual position to the interactive object with the smallest number of message interactions, assign the suboptimal virtual position to the interactive object with the second smallest number of message interactions, and so on, thereby assigning a virtual position to each interactive object and using the virtual position as the virtual interaction position, and so on.

[0143] In some embodiments, the terminal may display an interaction position configuration page, which includes at least one candidate interaction position area corresponding to a preset virtual interaction position and a configuration control corresponding to an interaction object, and in response to a move operation on the configuration control, a target interaction position area corresponding to each interaction object is determined in the candidate interaction position area, and interaction position configuration information of the target object in the target group is generated based on the target interaction position area. Based on the interaction position configuration information, a virtual interaction position corresponding to each interaction object is screened out in the preset virtual interaction position.

[0144] 205. The terminal converts the target voice message into virtual stereo based on the virtual interaction position.

[0145] For example, the terminal can obtain a set of azimuth channel data corresponding to at least one preset sound playback direction. The target sound playback direction corresponding to the virtual interaction position is queried in the preset channel playback direction. When the target sound playback direction exists, the azimuth channel data corresponding to the target sound playback direction is filtered out in the azimuth channel data set to obtain the target azimuth channel data. When the target sound playback direction does not exist, the azimuth distance between the target sound direction and each preset sound playback direction is calculated. Based on the azimuth distance, the preset sound playback directions are sorted. Based on the sorting result, the preset sound playback directions with the smallest azimuth distance are filtered out in the preset sound playback directions, thereby obtaining the adjacent sound playback direction of the target sound playback direction. The azimuth channel data corresponding to the adjacent sound playback direction is extracted from the azimuth channel data to obtain the target azimuth channel data.

[0146] The terminal convolves the target voice message with the target azimuth channel data to obtain a dual-channel stereo signal, and uses the stereo signal as a virtual stereo message, which can be specifically shown in formula (1).

[0147] 206. The terminal plays the virtual stereo according to the play order.

[0148] For example, the terminal can determine the target playback order of the virtual stereo message based on the playback order. According to the target playback order, the playback component is called to play the virtual stereo message, or the target playback order and the virtual stereo message can be sent to the object terminal corresponding to the target object so that the object terminal plays the virtual stereo message according to the target playback order, and so on.

[0149] As can be seen from the above, in the embodiment of the present application, after obtaining at least one to-be-played group message corresponding to the target object in the target group, the terminal identifies the play order and message type of the to-be-played group message in the target group, and then, according to the message type, performs voice conversion on the to-be-played group message to obtain a monophonic target voice message, determines the virtual interaction position corresponding to each interactive object in the preset virtual interaction position, and then, based on the virtual interaction position, converts the target voice message into a virtual stereo message, and plays the virtual stereo message according to the play order; since the scheme can perform voice conversion on the to-be-played group message according to the message type, it can realize the conversion of group messages of various message types into voice messages, and it can also determine the virtual interaction position of the interactive object interacting with the target object, so as to convert the converted monophonic target voice message into a virtual stereo message with orientation, and then distinguish different message senders (interaction objects) through the virtual stereo message, so that the message interaction in the target group can be closer to the real communication scene, and therefore, it can improve the reading efficiency of message reading, enhance the sense of presence and fun of message reading, and improve the interaction efficiency of message interaction in the group.

[0150] In order to better implement the above method, an embodiment of the present invention further provides a message reading device, which can be integrated into an electronic device, such as a server or a terminal. The terminal can include a tablet computer, a laptop computer and / or a personal computer.

[0151] For example, Fig.10 As shown, the message reading device may include an acquisition unit 301, an identification unit 302, a conversion unit 303, a determination unit 304 and a playback unit 305, as follows:

[0152] (1) Acquisition unit 301;

[0153] The acquisition unit 301 is configured to acquire at least one to-be-played group message corresponding to a target object, where the to-be-played group message includes an interaction message between at least one interaction object in the target group and the target object.

[0154] For example, the acquisition unit 301 may be specifically configured to filter out at least one group message to be played from the group messages received by the target object in the target group.

[0155] (2) identification unit 302;

[0156] The identification unit 302 is used to identify the play order and message type of the group messages to be played in the target group.

[0157] For example, the identification unit 302 can be specifically used to obtain message attribute information of the group message to be played, identify the message format and receiving time of the group message in the message attribute information, determine the message type of the group message to be played based on the message format, and sort the group messages to be played according to the receiving time, and determine the playing order of each group message to be played based on the sorting result.

[0158] (3) conversion unit 303;

[0159] The conversion unit 303 is used to perform voice conversion on the group message to be played according to the message type to obtain a monophonic target voice message.

[0160] For example, the conversion unit 303 can be specifically used to, when the message type of the group message to be played is a voice message, use the group message to be played as a mono target voice message; when the message type of the group message to be played is other types except voice messages, perform voice conversion on the group message to be played to obtain a mono target voice message.

[0161] (4) determining unit 304;

[0162] The determining unit 304 is used to determine a virtual interaction position corresponding to each interaction object in the preset virtual interaction position, where the virtual interaction position indicates a sound playing direction of an interaction message corresponding to the interaction object.

[0163] For example, the determination unit 304 may be specifically configured to count the number of interaction messages of each interaction object in the target group, sort the interaction objects according to the number of interaction messages, and select a virtual interaction position of each interaction object from the preset interaction positions based on the sorting result.

[0164] (5) Playback unit 305;

[0165] The playing unit 305 is used to convert the target voice message into a virtual stereo message based on the virtual interaction position, and play the virtual stereo message according to the playing order.

[0166] For example, the playback unit 305 can be specifically used to obtain an azimuth channel data set corresponding to at least one preset sound playback orientation, filter out target azimuth channel data corresponding to a virtual interaction position in the azimuth channel data set, fuse the target azimuth channel data with the target voice message to obtain a virtual stereo message, determine the target playback order of the virtual stereo message based on the playback order, and play the virtual stereo message according to the target playback order.

[0167] In specific implementation, the above units can be implemented as independent entities, or can be arbitrarily combined to be implemented as the same or several entities. The specific implementation of the above units can refer to the previous method embodiments, which will not be repeated here.

[0168] As can be seen from the above, in this embodiment, after the acquisition unit 301 acquires at least one to-be-played group message corresponding to the target object in the target group, the identification unit 302 identifies the play order and message type of the to-be-played group message in the target group, and then the conversion unit 303 performs voice conversion on the to-be-played group message according to the message type to obtain a monophonic target voice message, and the determination unit 304 determines the virtual interaction position corresponding to each interactive object in the preset virtual interaction position, and then the playback unit 305 converts the target voice message into a virtual stereo message based on the virtual interaction position, and plays the virtual stereo message according to the play order; because the scheme can perform voice conversion on the to-be-played group message according to the message type, thereby realizing the conversion of group messages of various message types into voice messages, and further, the virtual interaction position of the interactive object interacting with the target object can be determined, so as to convert the converted monophonic target voice message into a virtual stereo message with orientation, and then distinguish different message senders (interaction objects) through the virtual stereo message, so that the message interaction in the target group can be closer to the real communication scene, and therefore, the reading efficiency of the message reading can be improved.

[0169] The embodiment of the present invention further provides an electronic device, such as Fig.11 As shown, it shows a schematic diagram of the structure of an electronic device involved in an embodiment of the present invention, specifically:

[0170] The electronic device may include components such as a processor 401 with one or more processing cores, a memory 402 with one or more computer-readable storage media, a power supply 403, and an input unit 404. Those skilled in the art will appreciate that Fig.11 The electronic device structure shown in the figure does not constitute a limitation on the electronic device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0171] The processor 401 is the control center of the electronic device, and uses various interfaces and lines to connect various parts of the entire electronic device. By running or executing software programs and / or modules stored in the memory 402, and calling data stored in the memory 402, the processor 401 performs various functions of the electronic device and processes data. Optionally, the processor 401 may include one or more processing cores; preferably, the processor 401 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application programs, and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 401.

[0172] The memory 402 can be used to store software programs and modules. The processor 401 executes various functional applications and data processing by running the software programs and modules stored in the memory 402. The memory 402 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of the electronic device, etc. In addition, the memory 402 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the memory 402 may also include a memory controller to provide the processor 401 with access to the memory 402.

[0173] The electronic device also includes a power supply 403 for supplying power to each component. Preferably, the power supply 403 can be logically connected to the processor 401 through a power management system, so as to manage charging, discharging, power consumption and other functions through the power management system. The power supply 403 can also include one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, power status indicators and other arbitrary components.

[0174] The electronic device may further include an input unit 404, which may be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control.

[0175] Although not shown, the electronic device may further include a display unit, etc., which will not be described in detail herein. Specifically in this embodiment, the processor 401 in the electronic device will load the executable files corresponding to the processes of one or more application programs into the memory 402 according to the following instructions, and the processor 401 will run the application programs stored in the memory 402, thereby realizing various functions, as follows:

[0176] At least one to-be-played group message corresponding to the target object is obtained, the to-be-played group message includes an interactive message between at least one interactive object in the target group and the target object, the playback order and message type of the to-be-played group messages are identified in the target group, voice conversion is performed on the to-be-played group messages according to the message type to obtain a mono target voice message, a virtual interactive position corresponding to each interactive object is determined in a preset virtual interactive position, the virtual interactive position indicates a sound playback direction of the interactive message corresponding to the interactive object, based on the virtual interactive position, the target voice message is converted into a virtual stereo message, and the virtual stereo message is played according to the playback order.

[0177] For example, after the electronic device selects at least one group message to be played from the group messages received by the target object in the target group, it can obtain the message attribute information of the group message to be played, identify the message format and receiving time of the group message in the message attribute information, determine the message type of the group message to be played based on the message format, and sort the group messages to be played according to the receiving time, and determine the play order of each group message to be played based on the sorting result. When the message type of the group message to be played is a voice message, the group message to be played is used as a mono target voice message. When the message type of the group message to be played is other types other than voice messages, the group message to be played is voice-converted to obtain a mono target voice message. The number of interactive messages of each interactive object in the target group is counted, the interactive objects are sorted according to the number of interactive messages, and the virtual interactive position of each interactive object is selected from the preset interactive positions based on the sorting result. Obtain an azimuth channel data set corresponding to at least one preset sound playback orientation, filter out target azimuth channel data corresponding to a virtual interaction position from the azimuth channel data set, fuse the target azimuth channel data with a target voice message to obtain a virtual stereo message, determine a target playback order of the virtual stereo message based on the playback order, and play the virtual stereo message according to the target playback order.

[0178] The specific implementation of the above operations can be found in the previous embodiments and will not be described in detail here.

[0179] As can be seen from the above, after obtaining at least one group message to be played corresponding to the target object in the target group, the embodiment of the present application identifies the play order and message type of the group message to be played in the target group, and then, according to the message type, performs voice conversion on the group message to be played to obtain a monophonic target voice message, determines the virtual interaction position corresponding to each interactive object in the preset virtual interaction position, and then, based on the virtual interaction position, converts the target voice message into a virtual stereo message, and plays the virtual stereo message according to the play order; because the scheme can perform voice conversion on the group message to be played according to the message type, thereby realizing the conversion of group messages of various message types into voice messages, and further, the virtual interaction position of the interactive object interacting with the target object can be determined, so that the converted monophonic target voice message is converted into a virtual stereo message with orientation, and then different message senders (interaction objects) are distinguished through the virtual stereo message, so that the message interaction in the target group can be closer to the real communication scene, and therefore, the reading efficiency of message reading can be improved, the sense of presence and fun of message reading can be enhanced, and the interaction efficiency of message interaction in the group can be improved.

[0180] A person of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0181] To this end, an embodiment of the present invention provides a computer-readable storage medium, in which a plurality of instructions are stored, and the instructions can be loaded by a processor to execute the steps in any message reading method provided in the embodiment of the present invention. For example, the instructions can execute the following steps:

[0182] At least one to-be-played group message corresponding to the target object is obtained, the to-be-played group message includes an interactive message between at least one interactive object in the target group and the target object, the playback order and message type of the to-be-played group messages are identified in the target group, voice conversion is performed on the to-be-played group messages according to the message type to obtain a mono target voice message, a virtual interactive position corresponding to each interactive object is determined in a preset virtual interactive position, the virtual interactive position indicates a sound playback direction of the interactive message corresponding to the interactive object, based on the virtual interactive position, the target voice message is converted into a virtual stereo message, and the virtual stereo message is played according to the playback order.

[0183] For example, after at least one group message to be played is screened out from the group messages received by the target object in the target group, the message attribute information of the group message to be played is obtained, the message format and receiving time of the group message are identified in the message attribute information, the message type of the group message to be played is determined based on the message format, and the group messages to be played are sorted according to the receiving time, and the play order of each group message to be played is determined based on the sorting result. When the message type of the group message to be played is a voice message, the group message to be played is used as a mono target voice message, and when the message type of the group message to be played is other types except voice messages, the group message to be played is voice-converted to obtain a mono target voice message. The number of interactive messages of each interactive object in the target group is counted, and the interactive objects are sorted according to the number of interactive messages. Based on the sorting result, the virtual interactive position of each interactive object is screened out from the preset interactive positions. Obtain an azimuth channel data set corresponding to at least one preset sound playback orientation, filter out target azimuth channel data corresponding to a virtual interaction position from the azimuth channel data set, fuse the target azimuth channel data with a target voice message to obtain a virtual stereo message, determine a target playback order of the virtual stereo message based on the playback order, and play the virtual stereo message according to the target playback order.

[0184] The specific implementation of the above operations can be found in the previous embodiments, which will not be described in detail here.

[0185] The computer-readable storage medium may include: a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0186] Since the instructions stored in the computer-readable storage medium can execute the steps in any message reading method provided in the embodiments of the present invention, the beneficial effects that can be achieved by any message reading method provided in the embodiments of the present invention can be achieved. Please refer to the previous embodiments for details and will not be repeated here.

[0187] According to one aspect of the present application, a computer program product or a computer program is provided, the computer program product or the computer program including computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the electronic device executes the methods provided in various optional implementations of the above-mentioned message reading aspect or group message playing aspect.

[0188] The message reading method, device, electronic device and computer-readable storage medium provided in the embodiments of the present invention are introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for technical personnel in this field, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A message reading method, It is characterized in that include: Acquire at least one to-be-played group message corresponding to the target object, wherein the to-be-played group message includes an interaction message between at least one interaction object in the target group and the target object; Identifying the play order and message type of the to-be-played group message in the target group; Performing voice conversion on the to-be-played group message according to the message type to obtain a monophonic target voice message; Determining a virtual interaction position corresponding to each interaction object in the preset virtual interaction position, wherein the virtual interaction position indicates a sound playing direction of an interaction message corresponding to the interaction object; Based on the virtual interaction position, the target voice message is converted into a virtual stereo message, and the virtual stereo message is played according to the playing order.

2. The message reading method according to claim 1, It is characterized in that The step of determining a virtual interaction position corresponding to each interaction object in the preset virtual interaction position includes: Counting the number of interaction messages of each interaction object in the target group, where the number of interaction messages indicates the activity level of the interaction object in the target group; Based on the number of interaction messages, a virtual interaction position corresponding to each interaction object is determined in preset virtual interaction positions.

3. The message reading method according to claim 2, It is characterized in that The counting of the number of interaction messages of each interaction object in the target group includes: Filtering out a target group message to be played corresponding to each interactive object from the group messages to be played; The number of the target group messages to be played is counted in the group messages to be played, and the number of interaction messages of the interaction object is obtained.

4. The message reading method according to claim 2, It is characterized in that The counting of the number of interaction messages of each interaction object in the target group includes: Obtain at least one historical group message within a preset historical time range in the target group; The target historical group messages corresponding to each interaction object are screened out from the historical group messages to obtain the number of interaction messages of each interaction object.

5. The message reading method according to claim 2, It is characterized in that The step of determining a virtual interaction position corresponding to each interaction object in a preset virtual interaction position based on the number of interaction messages includes: sorting the interaction objects according to the number of the interaction messages; Based on the sorting result, a virtual interaction position corresponding to each interaction object is screened out from the preset interaction positions.

6. The message reading method according to claim 1, It is characterized in that The step of determining a virtual interaction position corresponding to each interaction object in the preset virtual interaction position includes: Acquire interaction position configuration information of the target object in the target group; Based on the interaction position configuration information, a virtual interaction position corresponding to each interaction object is screened out from the preset virtual interaction positions.

7. The message reading method according to claim 6, It is characterized in that The acquiring the interaction position configuration information of the target object in the target group includes: Displaying an interaction position configuration page, the interaction position configuration page including at least one candidate interaction position area corresponding to a preset virtual interaction position and a configuration control corresponding to the interaction object; In response to a move operation on the configuration control, determining a target interaction position area corresponding to each interaction object in the candidate interaction position area; Based on the target interaction position area, interaction position configuration information of the target object in the target group is generated.

8. The message reading method according to claim 1, It is characterized in that The converting the target voice message into a virtual stereo message based on the virtual interaction position includes: Obtaining a set of azimuth channel data corresponding to at least one preset sound playback orientation; Filtering out target azimuth channel data corresponding to the virtual interaction position from the azimuth channel data set; The target azimuth channel data is fused with the target voice message to obtain a virtual stereo message.

9. The message reading method according to claim 8, It is characterized in that The step of selecting target azimuth channel data corresponding to the virtual interaction position from the azimuth channel data set includes: Querying the target sound playing direction corresponding to the virtual interaction position in the preset sound playing direction; When the target sound playback direction exists, the azimuth channel data corresponding to the target sound playback direction is selected from the azimuth channel data set to obtain the target azimuth channel data; When the target sound playback direction does not exist, the adjacent sound playback direction of the target sound playback direction is determined in the preset sound playback direction, and the azimuth channel data corresponding to the adjacent sound playback direction is extracted from the azimuth channel data set to obtain the target azimuth channel data.

10. The message reading method according to claim 1, It is characterized in that The voice conversion of the to-be-played group message according to the message type to obtain a monophonic target voice message includes: When the message type of the group message to be played is a voice message, the group message to be played is used as a mono target voice message; When the message type of the group message to be played is other types except the voice message, voice conversion is performed on the group message to be played to obtain a monophonic target voice message.

11. The message reading method according to claim 10, It is characterized in that The voice conversion of the to-be-played group message to obtain a monophonic target voice message includes: Determining, according to the message type, a message conversion text of the group message to be played; A mono voice corresponding to the to-be-played group message is constructed based on the message conversion text to obtain a target voice message.

12. The message reading method according to claim 11, It is characterized in that The step of determining the message conversion text of the to-be-played group message according to the message type includes: When the message type is a text message, extracting text content from the to-be-played group message to obtain a message conversion text; When the message type is an image message, identifying a message conversion text in the to-be-played group message; When the message type is an emoticon message, an interaction scenario is identified in the target group, and based on the interaction scenario, the text corresponding to the to-be-played group message is screened out from a preset text set to obtain a message conversion text.

13. The message reading method according to claim 11, It is characterized in that The step of constructing a monophonic voice corresponding to the to-be-played group message based on the message conversion text to obtain a target voice message includes: Based on the object identifier of the interactive object, an acoustic feature is extracted from the message conversion text using a speech conversion model; The acoustic features are converted into monophonic voice to obtain a target voice message corresponding to the group message to be played.

14. The message reading method according to claim 13, It is characterized in that Before extracting the speech features corresponding to the interactive object from the message conversion text using the speech conversion model based on the object identification of the interactive object, the method further includes: Obtaining at least one historical voice message sample of the interactive object; Identifying text content in the historical voice message sample to obtain a message conversion text sample corresponding to the historical voice message sample; The preset voice conversion model is trained by using the historical voice message samples and the corresponding message conversion text samples to obtain a trained voice conversion model.

15. The message reading method according to claim 11, It is characterized in that The step of constructing a monophonic voice corresponding to the to-be-played group message based on the message conversion text to obtain a target voice message includes: Based on the object identifier of the interactive object, a target voice conversion model corresponding to the interactive object is selected from the voice conversion models corresponding to at least one preset sound effect; The target voice conversion model is used to convert the message conversion text into monophonic voice to obtain a target voice message corresponding to the interactive object.

16. The message reading method according to claim 1, It is characterized in that The obtaining of at least one to-be-played group message corresponding to the target object includes: Displaying a message interaction page of the target object in the target group, the message interaction page including a list of group messages to be played received within a preset time range; In response to a selection operation on the list of group messages to be played, determining a starting group message to be played corresponding to the selection operation; Based on the initial to-be-played group message, at least one to-be-played group message corresponding to the target object is screened out from the to-be-played group message list.

17. A message reading device, It is characterized in that include: An acquisition unit, configured to acquire at least one to-be-played group message corresponding to a target object, wherein the to-be-played group message includes an interaction message between at least one interaction object in a target group and the target object; an identification unit, configured to identify the play order and message type of the to-be-played group message in the target group; A conversion unit, configured to perform voice conversion on the to-be-played group message according to the message type to obtain a monophonic target voice message; A determination unit, configured to determine a virtual interaction position corresponding to each interaction object in a preset virtual interaction position, wherein the virtual interaction position indicates a sound playing direction of an interaction message corresponding to the interaction object; The playing unit is used to convert the target voice message into a virtual stereo message based on the virtual interaction position, and play the virtual stereo message according to the playing order.

18. An electronic device, It is characterized in that It comprises a processor and a memory, wherein the memory stores an application program, and the processor is used to run the application program in the memory to execute the steps in the message reading method according to any one of claims 1 to 16.

19. A computer program product comprising a computer program / instructions, It is characterized in that When the computer program / instructions are executed by a processor, the steps in the message reading method according to any one of claims 1 to 16 are implemented.

20. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor to execute the steps in the message reading method according to any one of claims 1 to 16.

Images