Voice recording method, device, equipment, storage medium and program product

Abstract

The application discloses a voice recording method and device, equipment, storage medium and program product, and belongs to the technical field of data processing. The method comprises the following steps: obtaining N voice signals corresponding to N wearable devices of a first user at a processing terminal, the voice signals being recorded in a conversation process between the first user and a second user, and N being an integer greater than 1; determining a phase difference between any two voice signals according to the N voice signals; determining N first distances between the N wearable devices and the second user according to the phase difference between any two voice signals and a sound propagation speed; determining a first voice signal of the first user and a second voice signal of the second user from the N voice signals according to the N first distances; and performing voice-to-text conversion processing on the first voice signal and the second voice signal to generate a voice recording manuscript.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of data processing, and particularly relates to a voice recording method and device, equipment, a storage medium and a program product. BACKGROUND

[0002] In a conference or an interview process, a recording device can be used to record the conversation content, and after the conference or the interview ends, a voice recording is generated. Currently, the record is often generated based on sound features to analyze the participants in the recording file, and then converted into text according to the voice signal of the participants to generate the record. However, in the actual conference or interview process, various complex scenarios may occur, such as multiple participants speaking at the same time or a noisy environment, etc. At this time, the sound features will be disturbed, resulting in poor reliability of the voice recording. SUMMARY

[0003] The embodiments of the present application aim to provide a voice recording method, device, equipment, storage medium and program product, which can improve the reliability of voice recording.

[0004] In a first aspect, the embodiments of the present application provide a voice recording method, which comprises:

[0005] In a case where the processing terminal obtains N voice signals corresponding to N wearable devices of a first user, the voice signals are recorded in a conversation process between the first user and a second user, and N is an integer greater than 1, the phase difference between any two voice signals is determined according to the N voice signals;

[0006] The N first distances between the N wearable devices and the second user are determined according to the phase difference between any two voice signals and the sound propagation speed;

[0007] The first voice signal of the first user and the second voice signal of the second user are determined from the N voice signals according to the N first distances;

[0008] The first voice signal and the second voice signal are subjected to voice-to-text conversion processing to generate a voice recording manuscript.

[0009] In a second aspect, the embodiments of the present application provide a voice recording device, which comprises:

[0010] A first determination module is configured to, in a case where the processing terminal obtains N voice signals corresponding to N wearable devices of a first user, the voice signals are recorded in a conversation process between the first user and a second user, and N is an integer greater than 1, determine the phase difference between any two voice signals according to the N voice signals;

[0011] The second determining module is configured to determine N first distances between the N wearable devices and the second user according to the phase difference between any two of the N voice signals and the sound propagation speed.

[0012] The third determining module is configured to determine a first voice signal of the first user and a second voice signal of the second user from the N voice signals according to the N first distances.

[0013] The processing module is configured to perform voice-to-text conversion processing on the first voice signal and the second voice signal to generate a voice recording manuscript.

[0014] In a third aspect, an electronic device is provided, which includes a processor and a memory. The memory stores programs or instructions executable on the processor. When the programs or instructions are executed by the processor, the steps of the method according to the first aspect are implemented.

[0015] In a fourth aspect, a readable storage medium is provided. The readable storage medium stores programs or instructions. When the programs or instructions are executed by a processor, the steps of the method according to the first aspect are implemented.

[0016] In a fifth aspect, a chip is provided. The chip includes a processor and a communication interface. The communication interface and the processor are coupled. The processor is configured to execute programs or instructions to implement the method according to the first aspect.

[0017] In a sixth aspect, a computer program product is provided. The program product is stored in a storage medium. When the program product is executed by at least one processor, the method according to the first aspect is implemented.

[0018] In the embodiments of the present application, when the terminal obtains N voice signals corresponding to N wearable devices of the first user, the voice signals are recorded in the conversation between the first user and the second user, and N is an integer greater than 1, the phase difference between any two of the N voice signals is determined according to the N voice signals; the N first distances between the N wearable devices and the second user are determined according to the phase difference between any two of the N voice signals and the sound propagation speed; the first voice signal of the first user and the second voice signal of the second user are determined from the N voice signals according to the N first distances; and the voice-to-text conversion processing is performed on the first voice signal and the second voice signal to generate a voice recording manuscript.

[0019] In this way, the voice signals can be recorded by multiple wearable devices at the same time, the distance between the device and the user is calculated according to the phase difference and the sound propagation speed of different voice signals, the voice signals of the first user and the second user are distinguished based on the distance and the spatial analysis of the voice signals, and then the voice recording script can be generated through voice-to-text conversion. The voice signals of different users are distinguished without using sound characteristics, avoiding the problem that the voice signals of different users are distinguished disorderly due to the interference of the recording environment on the sound characteristics, and effectively improving the reliability of the voice recording script. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 is a flowchart of a voice recording method provided by an embodiment of the present application;

[0021] Figure 2 is a schematic diagram of an application scenario in the voice recording method provided by an embodiment of the present application;

[0022] Figure 3 is a schematic diagram of one specific scenario embodiment of the voice recording method provided by an embodiment of the present application;

[0023] Figure 4 is a structural schematic diagram of a voice recording device provided by an embodiment of the present application;

[0024] Figure 5 is a structural schematic diagram of an electronic device provided by an embodiment of the present application;

[0025] Figure 6 is a hardware structural schematic diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0026] The technical solutions in the embodiments of the present application will be described clearly below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art belong to the scope of protection of the present application.

[0027] The terms "first", "second", and the like in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first", "second", etc. are usually a category and do not limit the number of objects, for example, the first object can be one or more. In addition, "and / or" in the specification and claims indicates at least one of the connected objects, and the character " / ", generally indicates that the objects before and after are in an "or" relationship.

[0028] The voice recording method provided by the embodiments of the present application will be described in detail below in combination with the drawings, specific embodiments and application scenarios thereof.

[0029] Figure 1 is a flowchart of the voice recording method provided by the embodiments of the present application. The voice recording method can include:

[0030] In step 101, the processing terminal obtains N voice signals corresponding to N wearable devices of the first user, the voice signals are recorded during the conversation between the first user and the second user, and in the case that N is an integer greater than 1, the phase difference between any two voice signals is determined according to the N voice signals.

[0031] In step 101, the first user and the second user can use the N wearable devices of the first user to record their conversation content during the conversation. The wearable devices can include earphones, mobile phones, watches, notebook computers, recording pens, etc., and their selection and number can be set according to actual conditions, which are not specifically limited here.

[0032] In order to facilitate the description of the technical solutions of the embodiments of the present application, as shown in Figure 2 , the following will be described by taking the number of wearable devices as 3. The first user 202 can use 3 wearable devices such as earphones, mobile phones and watches to record the conversation content when interviewing the second user 201.

[0033] After the interview is over, the processor can obtain N voice signals corresponding to N wearable devices, for example, the voice signal D1 corresponding to the earphones, the voice signal D2 corresponding to the watch and the voice signal D3 corresponding to the mobile phone.

[0034] It can be understood that the processing terminal can be a cloud server, or any wearable device. In other words, the voice recording method can be implemented by the cloud server integrating N voice signals for processing, or by any wearable device integrating N voice signals for processing.

[0035] The phase difference T1 between the voice signal D1 and the voice signal D2, the phase difference T2 between the voice signal D2 and the voice signal D3, and the phase difference T3 between the voice signal D3 and the voice signal D1 can be calculated according to the voice signals obtained by different wearable devices, by calculating mutual information entropy, or based on time translation of voice signal recording.

[0036] In step 102, the N first distances between the N wearable devices and the second user are determined according to the phase difference between any two voice signals and the speed of sound propagation.

[0037] In step 102, the first distances between the different wearable devices and the second user are calculated according to the sound propagation speed v of sound in air and the phase difference T1, the phase difference T2 and the phase difference T3.

[0038] For example, as shown in the figure, the first distance between the earphone and the second user 201 can be equal to the product of the phase difference T1 and the sound propagation speed v, the first distance between the watch and the second user 201 can be equal to the product of the phase difference T2 and the sound propagation speed v, and the first distance between the mobile phone and the second user 201 can be equal to the product of the phase difference T3 and the sound propagation speed v. Figure 2 The first distances calculated in this way are L11, L12 and L13 respectively.

[0039] In step 103, the first speech signal of the first user and the second speech signal of the second user are determined from the N speech signals according to the N first distances.

[0040] In step 103, the signals in the speech signal D1, the speech signal D2 and the speech signal D3 that satisfy L11=T1*V, L12=T2*V and L13=T3*V can be determined as the first speech signal U1 of the first user, and the first speech signal in the speech signal D1, the speech signal D2 and the speech signal D3 can be removed, and the remaining signals can be determined as the second speech signal U2 of the second user.

[0041] In some examples, the first speech signal U1 of the first user can include the speech signal of the first user in the speech signal D1, the speech signal of the first user in the speech signal D2 and the speech signal of the first user in the speech signal D3 respectively. The second speech signal U2 of the second user can include the speech signal of the second user in the speech signal D1, the speech signal of the second user in the speech signal D2 and the speech signal of the second user in the speech signal D3 respectively. Subsequently, speech recording can be performed on the first speech signal and the second speech signal corresponding to each wearable device respectively, and the speech recording results of each wearable device are compared to finally generate a complete speech recording manuscript.

[0042]

[0043] ​In some examples, the first speech signal U1 of the first user can be obtained by merging the speech signal of the first user in the comparison speech signal D1, the speech signal of the first user in the speech signal D2, and the speech signal of the first user in the speech signal D3. The second speech signal U2 of the second user can also be obtained by merging the speech signal of the second user in the comparison speech signal D1, the speech signal of the second user in the speech signal D2, and the speech signal of the second user in the speech signal D3. The merging processing can be splicing the same speech signal and the respective different speech signals to obtain the merged speech signal.

[0044] In step 104, the first speech signal and the second speech signal are subjected to speech-to-text conversion processing to generate a speech recording manuscript.

[0045] In step 104, the first speech signal and the second speech signal can be subjected to speech-to-text conversion processing respectively to obtain respective corresponding text information. The text information of different users has different annotations. The speech recording manuscript can be generated by splicing the text information with different annotations. For example, the first user can be annotated as TU1 and the second user can be annotated as TU2 in the speech recording manuscript, so that it can be clearly known who said each sentence in the speech recording manuscript.

[0046] In the embodiments of the present application, in the case that the processor obtains N speech signals corresponding to N wearable devices of the first user, the speech signals are recorded during the conversation between the first user and the second user, N is an integer greater than 1, the phase difference between any two speech signals is determined according to the N speech signals; the N first distances between the N wearable devices and the second user are determined according to the phase difference between any two speech signals and the speed of sound propagation; the first speech signal of the first user and the second speech signal of the second user are determined from the N speech signals according to the N first distances; and the speech-to-text conversion processing is performed on the first speech signal and the second speech signal to generate a speech recording manuscript.

[0047] In this way, multiple wearable devices can be used to record speech signals at the same time. The distance between the device and the user is calculated according to the phase difference and the speed of sound propagation of different speech signals. Based on the distance, the speech signals of the first user and the second user are distinguished through spatial analysis, and then the speech-to-text conversion is performed to generate a speech recording manuscript. The speech signals of different users do not need to be distinguished by sound characteristics, which avoids the problem that the sound characteristics are disturbed by the recording environment, resulting in confusion in distinguishing the speech signals of different users, thereby effectively improving the reliability of the speech recording manuscript.

[0048] In some embodiments, the above step 103 can include:

[0049] According to the N first distances, N third speech signals of the first user and N fourth speech signals of the second user are determined from the N speech signals respectively;

[0050] According to the N fourth speech signals, a phase difference between any two fourth speech signals and signal strengths corresponding to the N fourth speech signals are determined;

[0051] According to the phase difference between any two fourth speech signals and the signal strengths corresponding to the N fourth speech signals, N second distances between the N wearable devices and the second user are determined;

[0052] According to the N first distances and the N second distances, N third distances are determined;

[0053] According to the N third distances, a first speech signal of the first user and a second speech signal of the second user are determined from the N speech signals.

[0054] In this embodiment, in order to more accurately determine the distances between different wearable devices and the second user, so that the speech signals of different users can be more accurately distinguished subsequently, and the reliability of the speech recording script is improved, N third speech signals of the first user and N fourth speech signals of the second user can be determined from the N speech signals respectively according to the N first distances.

[0055] In other words, the signals in the speech signal D1, the speech signal D2 and the speech signal D3 that satisfy L11=T1*V, L12=T2*V and L13=T3*V can be determined as the third speech signals of the first user respectively, the third speech signals are removed from the speech signal D1, the speech signal D2 and the speech signal D3 respectively, and the remaining other part of the speech signals are determined as the fourth speech signals of the second user respectively.

[0056] The phase difference calculation is performed on any two fourth speech signals to obtain the phase difference T4 between the fourth speech signal in the speech signal D1 and the fourth speech signal in the speech signal D2, the phase difference T5 between the fourth speech signal in the speech signal D2 and the fourth speech signal in the speech signal D3, and the phase difference T3 between the fourth speech signal in the speech signal D3 and the fourth speech signal in the speech signal D1.

[0057] The signal strength N4 of the fourth speech signal in the speech signal D1, the signal strength N5 of the fourth speech signal in the speech signal D2, and the signal strength N6 of the fourth speech signal in the speech signal D3 can also be obtained.

[0058] It can be understood that the signal strength, the phase difference and the second distance between the wearable device and the second user can satisfy the following conditions:

[0059] L21-L22=T4*V

[0060] L22-L23=T5*V

[0061] L23-L21=T6*V

[0062] L21 / L23=N4 / N6

[0063] L22 / L23=N5 / N6

[0064] L21 / L22=N4 / N5

[0065] Based on the above conditions, the second distances L21, L22 and L23 between different wearable devices and the second user can be calculated.

[0066] The third distance L1 can be determined according to the first distance L11 and the second distance L21, the third distance L2 can be determined according to the first distance L12 and the second distance L22, and the third distance L3 can be determined according to the first distance L13 and the second distance L23. It can be understood that when recording the conversation content using the wearable device, if the recording environment at that time is relatively quiet and the conversation is smooth, the possibility of the voice signal being disturbed is small at this time, and in theory, the first distance and the second distance between the same wearable device and the second user are equal, and at this time, the first distance or the second distance is directly determined as the third distance.

[0067] If the voice signal is disturbed, the first distance and the second distance between the same wearable device and the second user calculated may be different, and at this time, the third distance can be calculated according to the weighted average method, and the third distance is taken as the final distance between the wearable device and the second user. For example, L1=(L11+L21) / 2, L2=(L12+L22) / 2 and L3=(L13+L23) / 2.

[0068] The first voice signal of the first user and the second voice signal of the second user can be determined from the N voice signals according to the N third distances. For example, the signal in the voice signal D1, the voice signal D2 and the voice signal D3 that satisfies L1=T1*V, L2=T2*V and L3=T3*V can be determined as the first voice signal U1 of the first user, and the first voice signal in the voice signal D1, the voice signal D2 and the voice signal D3 can be removed, and the remaining signal can be determined as the second voice signal U2 of the second user.

[0069] In this way, the distance between the different wearable devices and the second user can be calculated more accurately based on both the phase difference and the signal strength, has a certain degree of anti-interference capability, lays a foundation for spatial analysis of the voice signal, can more accurately distinguish the voice signals of the first user and the second user, and further improves the reliability of the voice recording script.

[0070] In some embodiments, after step 103, the voice recording method can further include:

[0071] According to the N voice signals, M fifth voice signals corresponding to M signal sources are determined, M being an integer greater than 2;

[0072] The first voice signal and the second voice signal are matched with the M fifth voice signals respectively, to determine a sixth voice signal corresponding to the first user and a seventh voice signal corresponding to the second user;

[0073] The above step 104 can include:

[0074] The sixth voice signal and the seventh voice signal are subjected to voice-to-text conversion processing to generate a voice recording script.

[0075] In the present embodiment, in the case where the recording environment is relatively complex, for example, in addition to the first user and the second user, other people or devices also emit sound and are recorded by the wearable device, in order to more accurately determine the voice signals of the first user and the second user, the voice information D1, the voice information D2 and the voice information D3 can also be separated based on the signal sources emitting the sound.

[0076] For example, the voice information D1, the voice information D2 and the voice information D3 can be integrated, and independent component analysis (ICA) processing is performed on the set to generate fifth voice signals corresponding to the signal sources, which can be denoted as voice signal S1, voice signal S2, voice signal S3, …, and voice signal SM.

[0077] It can be understood that the first voice signal can include the voice signal of the first user and the interference voice signal, and the second voice signal can include the voice signal of the second user and the interference voice signal. Then, the first voice signal U1 and the second voice signal U2 can be matched with the M fifth voice signals respectively. If the first voice signal U1 includes the voice signal S1, the voice signal S3,..., and the voice signal SM, and the second voice signal U2 includes the voice signal S2, the voice signal S3,..., and the voice signal SM, the voice signal S3 to the voice signal SM can be determined as the interference voice signal, the voice signal S1 can be determined as the sixth voice signal corresponding to the first user, and the voice signal S2 can be determined as the seventh voice signal corresponding to the second user.

[0078] In this way, the first voice signal and the second voice signal can be matched based on the fifth voice signals corresponding to different signal sources, the interference voice signals in the first voice signal and the second voice signal can be removed, and more accurate voice signals corresponding to the first user and the second user respectively can be obtained, so that the reliability of the voice recording document is further improved.

[0079] In some embodiments, the step 104 described above can include:

[0080] The first voice signal is subjected to voice-to-text conversion processing to obtain a first document, the first document includes a plurality of first sentences, and each first sentence carries time information;

[0081] The second voice signal is subjected to voice-to-text conversion processing to obtain a second document, the second document includes a plurality of second sentences, and each second sentence carries time information;

[0082] The first document and the second document are spliced in time sequence according to the time information carried by each first sentence and the time information carried by each second sentence to generate a voice recording document, wherein in the voice recording document, the first sentence is marked with an identity identifier of the first user, and the second sentence is marked with an identity identifier of the second user.

[0083] In this embodiment, the first voice signal can be subjected to voice-to-text conversion processing to obtain a first document TU1. It can be understood that the recorded voice signal generally has a time stamp. Based on the time stamp, the first document translated from the first voice signal can include a plurality of first sentences, and each first sentence carries time information. Similarly, the second voice signal can be subjected to voice-to-text conversion processing to obtain a second document TU2, and the second document can also include a plurality of second sentences, and each second sentence carries time information.

[0084] The first script and the second script can be spliced in chronological order according to the time information carried by each first sentence and the time information carried by each second sentence to generate the voice recording script TU.

[0085] In the voice recording script TU, the first user can be marked with an identity identifier, and the second user can be marked with an identity identifier. For example, TU1 and TU2 can be used to mark the dialogues in the voice recording script TU, that is, the first user and the second user.

[0086] In this way, the voice recording script is arranged in chronological order, and the words of the first user and the second user are marked respectively, so that subsequent viewing of the voice recording script can quickly understand the conversation content and recording situation, and the reliability of the voice recording script is improved.

[0087] In some embodiments, the above-mentioned splicing the first script and the second script in chronological order according to the time information carried by each first sentence and the time information carried by each second sentence to generate the voice recording script can further include:

[0088] Splicing the first script and the second script in chronological order according to the time information carried by each first sentence and the time information carried by each second sentence to generate a third script;

[0089] The third script is processed to obtain the voice recording script, and the first processing includes at least one of the following:

[0090] Correcting errors in the third script;

[0091] Correcting the sentences in the third script so that the corrected third script is consistent in nouns and smooth in sentences;

[0092] Adding a summary text paragraph, and the summary text paragraph is generated based on the text content of the third script.

[0093] In this embodiment, after the first script and the second script are spliced in chronological order according to the time information carried by each first sentence and the time information carried by each second sentence to generate a third script, the third script can be input into an artificial intelligence model. The artificial intelligence model can be any general artificial intelligence system that can effectively distinguish and generate language context, such as Chatgpt.

[0094] The artificial intelligence model can be used to correct the errors in the third manuscript. Alternatively, the sentences in the third manuscript can be corrected to make the corrected third manuscript consistent in terms of nouns and fluent in terms of sentences. It can be understood that the conversation content of the first user and the second user should be around the same matter, so the nouns used in their respective sentences should be consistent, and the conversation content should be continuous and contextually consistent.

[0095] Alternatively, a summary text paragraph can be generated for the text content of the third manuscript, and the summary text paragraph can be added to the third manuscript.

[0096] After processing the third manuscript using the artificial intelligence model, the final speech recording manuscript with more accurate expressions can be output, further improving the reliability of the speech recording manuscript.

[0097] As shown in Figure 3 , the present application embodiment also provides a specific scene embodiment of the above-mentioned speech recording method, which includes the following steps:

[0098] Step 301, conducting an interview and recording the conversation content using a plurality of wearable devices;

[0099] Step 302, performing spatial analysis according to the speech signals of the plurality of wearable devices to obtain a first speech signal of the interviewer and a second speech signal of the interviewee;

[0100] Step 303, excluding interference signals in the first speech signal and the second speech signal according to the fifth speech signal of the plurality of signal sources to obtain a sixth speech signal of the interviewer and a seventh speech signal of the interviewee;

[0101] Step 304, performing speech-to-text conversion processing on the sixth speech signal and the seventh speech signal;

[0102] Step 305, using an artificial intelligence model to correct and summarize the converted manuscript to obtain a speech recording manuscript.

[0103] The speech recording method provided by the present application embodiment can be executed by a speech recording device. In the present application embodiment, the method of the speech recording device for executing speech recording is taken as an example to illustrate the speech recording device provided by the present application embodiment.

[0104] As shown in Figure 4 , the speech recording device 400 provided by the present application embodiment can include:

[0105] The first determining module 401 is configured to, when the processing terminal obtains N voice signals corresponding to N wearable devices of the first user, the voice signals being recorded in a conversation process between the first user and the second user, and N being an integer greater than 1, determine a phase difference between any two voice signals according to the N voice signals.

[0106] The second determining module 402 is configured to determine N first distances between the N wearable devices and the second user according to the phase difference between any two voice signals and a sound propagation speed.

[0107] The third determining module 403 is configured to determine a first voice signal of the first user and a second voice signal of the second user from the N voice signals according to the N first distances.

[0108] The processing module 404 is configured to perform voice-to-text conversion processing on the first voice signal and the second voice signal to generate a voice recording manuscript.

[0109] In this way, the voice signals can be recorded by using multiple wearable devices at the same time, the distances between the devices and the user can be calculated according to the phase difference between different voice signals and the sound propagation speed, the voice signals of the first user and the second user can be distinguished based on the spatial analysis of the voice signals based on the distances, and then the voice recording manuscript can be generated through voice-to-text conversion. The voice signals of different users do not need to be distinguished by using sound features, which avoids the problem that the voice signals of different users are distinguished disorderly due to the interference of the recording environment on the sound features, and effectively improves the reliability of the voice recording manuscript.

[0110] In some embodiments, the third determining module 403 can be further configured to:

[0111] determine N third voice signals of the first user and N fourth voice signals of the second user from the N voice signals respectively according to the N first distances;

[0112] determine a phase difference between any two fourth voice signals and signal intensities corresponding to the N fourth voice signals according to the N fourth voice signals;

[0113] determine N second distances between the N wearable devices and the second user according to the phase difference between any two fourth voice signals and the signal intensities corresponding to the N fourth voice signals;

[0114] determine N third distances according to the N first distances and the N second distances;

[0115] determine the first voice signal of the first user and the second voice signal of the second user from the N voice signals according to the N third distances.

[0116] In this way, the distance between the different wearable devices and the second user can be calculated more accurately based on both the phase difference and the signal strength, has a certain degree of anti-interference capability, lays a foundation for spatial analysis of the voice signal, can more accurately distinguish the voice signals of the first user and the second user, and further improves the reliability of the voice recording script.

[0117] In some embodiments, the voice recording device 400 can further include:

[0118] The fourth determination module is configured to determine M fifth voice signals corresponding to M signal sources according to the N voice signals, where M is an integer greater than 2.

[0119] The fifth determination module is configured to match the first voice signal and the second voice signal with the M fifth voice signals respectively, and determine a sixth voice signal corresponding to the first user and a seventh voice signal corresponding to the second user.

[0120] The processing module 404 can be further configured to:

[0121] The processing module 404 can be further configured to:

[0122] In this way, the distance between the different wearable devices and the second user can be calculated more accurately based on both the phase difference and the signal strength, has a certain degree of anti-interference capability, lays a foundation for spatial analysis of the voice signal, can more accurately distinguish the voice signals of the first user and the second user, and further improves the reliability of the voice recording script.

[0123] In some embodiments, the processing module 404 can be further configured to:

[0124] The processing module 404 can be further configured to:

[0125] The processing module 404 can be further configured to:

[0126] The processing module 404 can be further configured to:

[0127] In this way, the voice recording script is sorted in chronological order, and the words spoken by the first user and the second user are marked respectively, so that subsequent viewing of the voice recording script can quickly understand the conversation content and recording situation, and the reliability of the voice recording script is improved.

[0128] In some embodiments, the processing module 404 can also be configured to:

[0129] According to the time information carried by each first sentence and the time information carried by each second sentence, the first script and the second script are spliced in chronological order to generate a third script;

[0130] The third script is subjected to first processing to obtain a voice recording script, and the first processing includes at least one of the following:

[0131] Correcting errors in the third script;

[0132] Correcting the sentences in the third script so that the corrected third script is consistent in nouns and the sentences are smooth;

[0133] Adding a summary text paragraph, and the summary text paragraph is generated based on the text content of the third script.

[0134] After the third script is processed by the artificial intelligence model, a final voice recording script with more accurate expression can be output, further improving the reliability of the voice recording script.

[0135] The voice recording device in the embodiments of the present application can be an electronic device or a component in an electronic device, such as an integrated circuit or a chip. The electronic device can be a terminal or other devices other than a terminal. For example, the electronic device can be a mobile phone, a tablet computer, a notebook computer, a palm computer, a vehicle-mounted electronic device, a mobile Internet device (MID), an augmented reality (AR) / virtual reality (VR) device, a robot, a wearable device, an ultra-mobile personal computer (UMPC), a netbook, or a personal digital assistant (PDA), etc. The electronic device can also be a server, a network attached storage (NAS), a personal computer (PC), a television (TV), a teller machine, or a self-service machine, etc. The embodiments of the present application are not limited in this regard.

[0136] The voice recording device in the embodiments of the present application can be a device with an operating system. The operating system can be an Android operating system, an ios operating system, or other possible operating systems, which are not limited in the embodiments of the present application.

[0137] The voice recording device provided in the embodiments of the present application can realize Figures 1 to 3 The method embodiments realize various processes, which will not be repeated here to avoid repetition.

[0138] Optionally, as shown in Figure 5 The embodiments of the present application also provide an electronic device 500, which includes a processor 501 and a memory 502, and the memory 502 stores programs or instructions executable on the processor 501. When the programs or instructions are executed by the processor 501, each step of the above voice recording method embodiments is realized, and the same technical effects are achieved. To avoid repetition, each step will not be repeated here.

[0139] It should be noted that the electronic device in the embodiments of the present application includes the mobile electronic device and the non-mobile electronic device described above.

[0140] Figure 6 To realize the hardware structure of an electronic device in the embodiments of the present application.

[0141] The electronic device 600 includes but is not limited to a radio frequency unit 601, a network module 602, an audio output unit 603, an input unit 604, a sensor 605, a display unit 606, a user input unit 607, an interface unit 608, a memory 609, and a processor 610, etc.

[0142] Those skilled in the art can understand that the electronic device 600 can also include a power supply (such as a battery) for powering each component. The power supply can be logically connected to the processor 610 through a power management system, so as to realize functions such as management of charging, discharging, and power consumption management through the power management system. Figure 6 The electronic device structure shown in the embodiments of the present application does not constitute a limitation on the electronic device. The electronic device can include more or fewer components than those shown, or combine certain components, or arrange different components, which will not be repeated here.

[0143] The processor 610 can be configured to:

[0144] In a case where the terminal obtains N voice signals corresponding to N wearable devices of the first user, the voice signals are recorded in a conversation process between the first user and the second user, and N is an integer greater than 1, the phase difference between any two voice signals is determined according to the N voice signals.

[0145] determine N first distances between the N wearable devices and the second user according to the phase difference between any two of the N voice signals and the speed of sound propagation;

[0146] determine a first voice signal of the first user and a second voice signal of the second user from the N voice signals according to the N first distances;

[0147] perform voice-to-text conversion processing on the first voice signal and the second voice signal to generate a voice recording transcript.

[0148] In this way, multiple wearable devices can be used to record voice signals at the same time, and the distances between the devices and the users can be calculated according to the phase difference between different voice signals and the speed of sound propagation. Based on the distances, spatial analysis can be performed on the voice signals to distinguish the voice signals of the first user and the second user, respectively. Then, through voice-to-text conversion, a voice recording transcript can be generated. Without using sound features to distinguish the voice signals of different users, the problem of confusing the voice signals of different users due to interference of the recording environment on the sound features can be avoided, and the reliability of the voice recording transcript can be effectively improved.

[0149] In some embodiments, the processor 610 can be further configured to:

[0150] determine N third voice signals of the first user and N fourth voice signals of the second user from the N voice signals according to the N first distances, respectively;

[0151] determine a phase difference between any two of the N fourth voice signals and signal strengths corresponding to the N fourth voice signals according to the N fourth voice signals;

[0152] determine N second distances between the N wearable devices and the second user according to the phase difference between any two of the N fourth voice signals and the signal strengths corresponding to the N fourth voice signals;

[0153] determine N third distances according to the N first distances and the N second distances;

[0154] determine a first voice signal of the first user and a second voice signal of the second user from the N voice signals according to the N third distances.

[0155] In this way, more accurate distances between different wearable devices and the second user can be calculated based on both the phase difference and the signal strength, which has a certain degree of anti-interference capability. This lays a foundation for spatial analysis of the voice signals, and can more accurately distinguish the voice signals of the first user and the second user, further improving the reliability of the voice recording transcript.

[0156] In some embodiments, the processor 610 can be further configured to:

[0157] determine M fifth speech signals corresponding to the M signal sources according to the N speech signals, M being an integer greater than 2;

[0158] match the first speech signal and the second speech signal with the M fifth speech signals respectively, to determine a sixth speech signal corresponding to the first user and a seventh speech signal corresponding to the second user;

[0159] perform speech-to-text conversion processing on the sixth speech signal and the seventh speech signal to generate a speech recording script.

[0160] In this way, the first speech signal and the second speech signal can be matched based on the fifth speech signals corresponding to different signal sources, and the interference speech signals in the first speech signal and the second speech signal can be eliminated, so that more accurate speech signals corresponding to the first user and the second user respectively are obtained, thereby further improving the reliability of the speech recording script.

[0161] In some embodiments, the processor 610 can be further configured to:

[0162] perform speech-to-text conversion processing on the first speech signal to obtain a first script, the first script including a plurality of first sentences, and each first sentence carrying time information;

[0163] perform speech-to-text conversion processing on the second speech signal to obtain a second script, the second script including a plurality of second sentences, and each second sentence carrying time information;

[0164] splice the first script and the second script in chronological order according to the time information carried by each first sentence and the time information carried by each second sentence to generate a speech recording script, wherein in the speech recording script, the first sentences are marked with an identity identifier of the first user, and the second sentences are marked with an identity identifier of the second user.

[0165] In this way, the speech recording script is arranged in chronological order, and the words spoken by the first user and the second user are marked respectively, so that subsequent viewing of the speech recording script can quickly understand the conversation content and recording situation, thereby improving the reliability of the speech recording script.

[0166] In some embodiments, the processor 610 can be further configured to:

[0167] splice the first script and the second script in chronological order according to the time information carried by each first sentence and the time information carried by each second sentence to generate a third script;

[0168] perform first processing on the third script using an artificial intelligence model to obtain the speech recording script, the first processing including at least one of the following:

[0169] Correct the misspelled words in the third draft;

[0170] Correct the sentences in the third draft to make the nouns consistent and the sentences fluent in the corrected third draft;

[0171] Add a summary text paragraph, which is generated based on the text content of the third draft.

[0172] After processing the third draft using the artificial intelligence model, a final voice recording draft with more accurate expressions can be output, further improving the reliability of the voice recording draft.

[0173] It should be understood that in the embodiments of the present application, the input unit 604 can include a graphics processing unit (GPU) 6041 and a microphone 6042. The graphics processing unit 6041 processes image data of a still picture or a video obtained by an image capture device (such as a camera) in a video capture mode or an image capture mode. The display unit 606 can include a display panel 6061, which can be configured in the form of a liquid crystal display, an organic light-emitting diode, etc. The user input unit 607 includes at least one of a touch panel 6061 and other input devices 6072. The touch panel 6061 is also called a touch screen. The touch panel 6061 can include a touch detection device and a touch controller. The other input devices 6072 can include, but are not limited to, a physical keyboard, function keys (such as volume control keys, on-off keys, etc.), a trackball, a mouse, an operating lever, etc., which will not be described here.

[0174] The memory 609 can be used to store software programs and various data. The memory 609 can mainly include a first storage area storing programs or instructions and a second storage area storing data, wherein the first storage area can store an operating system, application programs or instructions required by at least one function (such as a sound playing function, an image playing function, etc.), and the like. In addition, the memory 609 can include a volatile memory or a non-volatile memory, or the memory 609 can include both a volatile memory and a non-volatile memory. The non-volatile memory can be a Read-Only Memory (ROM), a Programmable ROM (PROM), an Erasable PROM (EPROM), an Electrically EPROM (EEPROM), or a flash memory. The volatile memory can be a Random Access Memory (RAM), a Static RAM (SRAM), a Dynamic RAM (DRAM), a Synchronous DRAM (SDRAM), a Double Data Rate SDRAM (DDR SDRAM), an Enhanced SDRAM (ESDRAM), a Synch link DRAM (SLDRAM), and a Direct Rambus RAM (DRRAM). The memory 609 in the embodiments of the present application includes but is not limited to these and any other suitable types of memory.

[0175] The processor 610 can include one or more processing units; optionally, the processor 610 integrates an application processor and a modem processor, wherein the application processor mainly processes operations related to an operating system, a user interface, and an application program, and the modem processor mainly processes wireless communication signals, such as a baseband processor. It can be understood that the above-mentioned modem processor can also not be integrated into the processor 610.

[0176] The embodiments of the present application also provide a readable storage medium, the readable storage medium stores programs or instructions, the programs or instructions are executed by a processor to realize various processes of the above-mentioned voice recording method embodiments, and the same technical effects can be achieved. To avoid repetition, details are not described here.

[0177] The processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes a computer readable storage medium, such as a computer readable only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0178] The embodiment of the present application further provides a chip, which comprises a processor and a communication interface, the communication interface is coupled with the processor, the processor is used for running programs or instructions to realize the processes of the above voice recording method embodiments and achieve the same technical effects. To avoid repetition, details are not described herein.

[0179] It should be understood that the chip mentioned in the embodiment of the present application can also be referred to as a system level chip, a system chip, a chip system or a system on chip, etc.

[0180] The embodiment of the present application provides a computer program product, which is stored in a storage medium, and is executed by at least one processor to realize the processes of the above voice recording method embodiments and achieve the same technical effects. To avoid repetition, details are not described herein.

[0181] It should be noted that in this document, the term "comprising" or "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or device. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the method and device in the embodiment of the present application is not limited to the order of performing the functions as shown or discussed, but can also include performing the functions in a substantially simultaneous manner or in the opposite order, for example, the described method can be performed in an order different from that described, and various steps can also be added, omitted or combined. In addition, the features described with reference to some examples can be combined in other examples.

[0182] Through the above description of the embodiments, those skilled in the art can clearly understand that the above-mentioned example methods can be realized by means of software and a necessary general hardware platform, and of course, can also be realized by hardware, but in many cases, the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a computer software product in essence or in the form of a part that contributes to the prior art, which is stored in a storage medium (such as a ROM / RAM, a magnetic disc, an optical disc), and includes a plurality of instructions for causing a terminal (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in the various embodiments of the present application.

[0183] The embodiments of the present application are described above in combination with the drawings, but the present application is not limited to the above-mentioned specific embodiments, and the above-mentioned specific embodiments are only illustrative and not restrictive. Those skilled in the art can make many forms under the inspiration of the present application without departing from the scope of the present application and the scope protected by the claims.

Claims

1. A voice recording method, characterized by, The method comprises the following steps: In the case that the processing terminal acquires N voice signals corresponding to N wearable devices of a first user, the voice signals are recorded during a conversation between the first user and a second user, N is an integer greater than 1, and the phase difference between any two voice signals is determined according to the N voice signals; According to the phase difference between the two voice signals and the sound propagation speed, N first distances between the N wearable devices and the second user are determined; According to the N first distances, a first voice signal of the first user and a second voice signal of the second user are determined from the N voice signals; The first voice signal and the second voice signal are subjected to speech-to-text conversion processing to generate a voice recording manuscript; The first voice signal and the second voice signal are subjected to speech-to-text conversion processing to generate a voice recording manuscript, which comprises the following steps: According to the N first distances, N third voice signals of the first user and N fourth voice signals of the second user are determined from the N voice signals, respectively; According to the phase difference between any two fourth voice signals and the signal intensity corresponding to the N fourth voice signals, N second distances between the N wearable devices and the second user are determined; According to the N first distances and the N second distances, N third distances are determined; According to the N third distances, a first voice signal of the first user and a second voice signal of the second user are determined from the N voice signals. After the first voice signal and the second voice signal are determined from the N voice signals according to the N first distances, the method further comprises the following steps:

2. The method of claim 1, wherein, According to the N voice signals, M fifth voice signals corresponding to M signal sources are determined, M is an integer greater than 2; The first voice signal and the second voice signal are matched with the M fifth voice signals, respectively, to determine a sixth voice signal corresponding to the first user and a seventh voice signal corresponding to the second user; The first voice signal and the second voice signal are subjected to speech-to-text conversion processing to generate a voice recording manuscript, which comprises the following steps: The sixth voice signal and the seventh voice signal are subjected to speech-to-text conversion processing to generate a voice recording manuscript. The first voice signal and the second voice signal are subjected to speech-to-text conversion processing to generate a voice recording manuscript, which comprises the following steps:

3. The method of claim 1, wherein, The first voice signal is subjected to speech-to-text conversion processing to obtain a first manuscript, the first manuscript comprises a plurality of first sentences, and each first sentence carries time information; The second voice signal is subjected to speech-to-text conversion processing to obtain a second manuscript, the second manuscript comprises a plurality of second sentences, and each second sentence carries time information; ​ According to the time information carried by each first sentence and the time information carried by each second sentence, the first manuscript and the second manuscript are spliced in chronological order to generate a voice recording manuscript, wherein in the voice recording manuscript, the first sentence is marked with the identity of the first user, and the second sentence is marked with the identity of the second user.

4. The method of claim 3, wherein, According to the time information carried by each first sentence and the time information carried by each second sentence, the first manuscript and the second manuscript are spliced in chronological order to generate a voice recording manuscript, wherein in the voice recording manuscript, the first sentence is marked with the identity of the first user, and the second sentence is marked with the identity of the second user. According to the time information carried by each first sentence and the time information carried by each second sentence, the first manuscript and the second manuscript are spliced in chronological order to generate a third manuscript; The third manuscript is processed to obtain a voice recording manuscript, and the first processing includes at least one of the following: Correcting the wrong words in the third manuscript; Correcting the sentences in the third manuscript so that the corrected third manuscript is consistent in nouns and the sentences are smooth; Adding a summary text paragraph, which is generated based on the text content of the third manuscript.

5. A voice recording device, characterized by Comprise: The first determination module is used for obtaining N voice signals corresponding to N wearable devices of a first user, the voice signals being recorded during a conversation between the first user and a second user, and when N is an integer greater than 1, determining the phase difference between any two voice signals according to the N voice signals; The second determination module is used for determining N first distances between the N wearable devices and the second user according to the phase difference between any two voice signals and the speed of sound propagation; The third determination module is used for determining a first voice signal of the first user and a second voice signal of the second user from the N voice signals according to the N first distances; The processing module is used for performing voice-to-text conversion processing on the first voice signal and the second voice signal to generate a voice recording manuscript; The third determination module is used for determining a first voice signal of the first user and a second voice signal of the second user from the N voice signals according to the N first distances, comprising: According to the N first distances, N third voice signals of the first user and N fourth voice signals of the second user are determined from the N voice signals respectively; According to the phase difference between any two fourth voice signals and the signal intensity corresponding to the N fourth voice signals, N second distances between the N wearable devices and the second user are determined; According to the N first distances and the N second distances, N third distances are determined; According to the N third distances, the first voice signal of the first user and the second voice signal of the second user are determined from the N voice signals. The device further comprises:

6. The apparatus of claim 5, wherein, ​ a fourth determining module, configured to determine M fifth speech signals corresponding to M signal sources according to the N speech signals, M being an integer greater than 2; a fifth determining module, configured to match the first speech signal and the second speech signal respectively with the M fifth speech signals, to determine a sixth speech signal corresponding to the first user and a seventh speech signal corresponding to the second user; the speech-to-text conversion processing on the first speech signal and the second speech signal to generate a speech recording manuscript, comprising: the speech-to-text conversion processing on the sixth speech signal and the seventh speech signal to generate a speech recording manuscript.

7. The apparatus of claim 5, wherein, the speech-to-text conversion processing on the first speech signal and the second speech signal to generate a speech recording manuscript, comprising: performing speech-to-text conversion processing on the first speech signal to obtain a first manuscript, the first manuscript comprising a plurality of first sentences, and each first sentence carrying time information; performing speech-to-text conversion processing on the second speech signal to obtain a second manuscript, the second manuscript comprising a plurality of second sentences, and each second sentence carrying time information; performing speech-to-text conversion processing on the first speech signal and the second speech signal to generate a speech recording manuscript, comprising:

8. The apparatus of claim 7, wherein, performing speech-to-text conversion processing on the first speech signal and the second speech signal to generate a speech recording manuscript, comprising: performing speech-to-text conversion processing on the first speech signal and the second speech signal to generate a speech recording manuscript, comprising: performing speech-to-text conversion processing on the first speech signal and the second speech signal to generate a speech recording manuscript, comprising: performing speech-to-text conversion processing on the first speech signal and the second speech signal to generate a speech recording manuscript, comprising: a processor and a memory, the memory storing programs or instructions executable on the processor, the programs or instructions being executed by the processor to implement the steps of the method according to any one of claims 1-4. a readable storage medium storing programs or instructions, the programs or instructions being executed by a processor to implement the steps of the method according to any one of claims 1-4.

9. An electronic device, comprising:

11. A computer program product stored in a storage medium, the program product being executed by at least one processor to implement the steps of the method according to any one of claims 1-4.

10. A readable storage medium, characterized by, ​ ​

Images