Single-channel audio to text conversion method, device, electronic device and storage medium

Abstract

The present application discloses a method, device, electronic device and storage medium for converting single-channel audio into text, wherein the method comprises: obtaining the audio to be transcribed of the speech of the participants between two voice pauses in the conference scene; obtaining the role volume information of each participant in the audio to be transcribed; according to the volume information of each role, determining the target participant who theoretically outputs the audio to be transcribed among the participants; and transcribing the audio to be transcribed into an audio text with the identity of the target participant. The present application solves the technical problem that transcribing audio with multiple roles into an audio text with participant information requires high CPU performance.

Description

Technical Field

[0001] The present application relates to the field of audio processing technology, and in particular to a method, device, electronic device and storage medium for converting single-channel audio into text. Background Art

[0002] In various social activities, especially business activities, people are holding various meetings almost all the time. After the meeting, they need to make meeting minutes based on the content discussed in the meeting and the resolutions formed, and send them to the participants or archive them. At present, if the audio of multiple roles participating in the meeting is to be transcribed into text that can distinguish the speakers in real time, multiple channels are required to transcribe the audio of different participants simultaneously, which requires high CPU (central processing unit) performance. Summary of the invention

[0003] The main purpose of this application is to provide a single-channel audio-to-text method, device, electronic device and storage medium, aiming to solve the technical problem that transcribing audio with multiple roles into audio text with participant information requires high CPU performance.

[0004] To achieve the above object, the present application provides a single-channel audio-to-text method, the single-channel audio-to-text method comprising:

[0005] Obtain the audio to be transcribed of the participant's speech between two voice pauses in the conference scene;

[0006] Obtaining the role volume information of each participant in the audio to be transcribed;

[0007] According to the volume information of each role, determining a target participant from among all the participants to theoretically output the audio to be transcribed;

[0008] The audio to be transcribed is transcribed into an audio text having the identity identifier of the target participant.

[0009] Optionally, the step of obtaining the role volume information of each participant in the audio to be transcribed includes:

[0010] Decomposing the audio to be transcribed into the role audio of each participant;

[0011] The role volume information corresponding to each participant is obtained from each role audio respectively.

[0012] Optionally, the role volume information includes a segment volume value, and the step of respectively acquiring the role volume information corresponding to each participant from each role audio includes:

[0013] Taking a preset unit time as a time period, dividing each of the role audios in time sequence to obtain a sub-role audio group in each of the time periods;

[0014] Get the segment volume value of the sub-role audio of each participant in each of the sub-role audio groups.

[0015] Optionally, the step of determining a target participant who theoretically outputs the audio to be transcribed from among the participants according to the volume information of each role includes:

[0016] Compare the volume values ​​of all the clips corresponding to each time period respectively;

[0017] The participant corresponding to the sub-role audio with the largest clip volume value in each time period is determined as the time period role of this time period;

[0018] The time period role with the most repetitions is determined as the target participant of the audio to be transcribed.

[0019] Optionally, the step of transcribing the audio to be transcribed into an audio text having the identity identifier of the target participant includes:

[0020] Combine all sub-role audios corresponding to each time period into one audio package;

[0021] The audio packages are sequentially transcribed into audio segment texts in chronological order, and the audio segment texts are connected in the order of transcription to obtain an initial audio text;

[0022] The initial audio text is marked according to the target participant to obtain an audio text with the identity identifier of the target participant.

[0023] Optionally, after the step of transcribing the audio to be transcribed into an audio text having the identity identifier of the target participant, the method further includes:

[0024] Generating a time stamp of the audio text according to the reception time information of each of the audio segments to be transcribed;

[0025] The audio text is displayed according to the time mark by a display device.

[0026] Optionally, after the step of generating the time mark of the audio text according to the reception time information of each audio segment to be transcribed, the method further includes:

[0027] The audio text and a time stamp of the audio text are saved.

[0028] The present application also provides a single-channel audio-to-text device, which is applied to a single-channel audio-to-text device, and the single-channel audio-to-text device includes:

[0029] A receiving module, used to obtain the audio to be transcribed of the speech of the participant between two voice pauses in the conference scene;

[0030] An acquisition module, used to acquire the role volume information of each participant in the audio to be transcribed;

[0031] A target participant determination module, used to determine a target participant who theoretically outputs the audio to be transcribed from among the participants according to the volume information of each role;

[0032] The transcription module is used to transcribe the audio to be transcribed into an audio text having the identity identifier of the target participant.

[0033] The present application also provides an electronic device, which is a physical device, and includes: a memory, a processor, and a program of the single-channel audio-to-text method stored in the memory and executable on the processor. When the program of the single-channel audio-to-text method is executed by the processor, the steps of the single-channel audio-to-text method described above can be implemented.

[0034] The present application also provides a storage medium, which is a computer-readable storage medium, on which is stored a program for implementing a method for converting single-channel audio to text. When the program for the method for converting single-channel audio to text is executed by a processor, the steps of the method for converting single-channel audio to text as described above are implemented.

[0035] The present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned single-channel audio-to-text conversion method when executed by a processor.

[0036] In various social activities, especially business activities, people are holding various meetings almost all the time. After the meeting, they need to make meeting minutes based on the content discussed in the meeting and the resolutions formed, and send them to the participants or archive them. At present, if the audio of multiple roles participating in the meeting is to be transcribed into text that can distinguish the speakers in real time, multiple channels are required to transcribe the audio of different participants simultaneously, which requires high CPU (central processing unit) performance.

[0037] The present application provides a single-channel audio-to-text method, device, electronic device and storage medium, which obtains the audio to be transcribed of the speech of the participant between two voice pauses in the conference scene, obtains the role volume information of each participant in the audio to be transcribed, and realizes the acquisition of the role volume information of each participant in the audio to be transcribed with multiple participants. According to the role volume information, the amount of information output by each participant in the audio to be transcribed can be determined, and then by determining the target participant who theoretically outputs the audio to be transcribed from among the participants according to the role volume information, the target of theoretically outputting the audio to be transcribed in the audio to be transcribed is realized. The determination of participants, by transcribing the audio to be transcribed into an audio text with the identity identification of the target participant, realizes single-channel transcription of audio to be transcribed with the participation of multiple people, effectively reduces the requirements on the CPU of the user device, and judges the target participant who theoretically outputs the audio to be transcribed in the audio to be transcribed according to the role volume information, overcomes the problem that single-channel transcription cannot distinguish the participants to obtain the text with the participant information, realizes the audio text with accurate participant information through single-channel transcription, and overcomes the technical problem that the transcription of audio with the participation of multiple roles into audio text with the participant information has high CPU performance requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0039] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0040] Figure 1 This is a flowchart of an embodiment of a method for converting single-channel audio to text in the present application;

[0041] Figure 2 This is a flow chart of another embodiment of the method for converting single-channel audio to text of the present application;

[0042] Figure 3 This is a schematic diagram of a scenario of an embodiment of a method for converting single-channel audio to text in the present application;

[0043] Figure 4 This is a schematic diagram of the device structure of the hardware operating environment involved in the single-channel audio-to-text conversion method in the embodiment of the present application.

[0044] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0045] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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 ordinary technicians in this field without creative work belong to the scope of protection of the present invention.

[0046] The present application embodiment provides a single-channel audio-to-text method. In the first embodiment of the single-channel audio-to-text method of the present application, refer to Figure 1 , the single-channel audio-to-text method comprises:

[0047] Step S10, obtaining the audio to be transcribed of the participant's speech between two voice pauses in the conference scene;

[0048] In this embodiment, it should be noted that this embodiment is applied to a user device. In a conference scenario, the user device can collect sound signals emitted by each participant during a local meeting through a microphone, and can also receive audio signals sent by each end participating in the network conference through a network connection, and then implement audio processing, target participant determination, audio transcription and other functions through the functional modules of the user device itself. The functional modules are connected by wire harnesses without relying on the network.

[0049] Specifically, the audio to be transcribed of the participant's speech between two voice pauses in the conference scene is obtained in real time, wherein the voice pause is the time when the audio signal is interrupted during the conference. The voice pause can be set according to actual conditions or test results, for example, when the audio signals of two or more participants are detected to alternate, it may be that the participant has changed, when the audio signal is detected to suddenly decrease to a preset intensity, etc. It is easy to understand that the time when the complete conference audio starts is taken as the first voice pause, and when the voice pause is detected for the first time after the start of the audio to be transcribed, it is taken as the second voice pause, and the audio between the first voice pause and the second voice pause is also a section of audio to be transcribed. It is easy to understand that the audio to be transcribed can be the user device collecting the sound signals emitted by each participant during the local meeting through a microphone and then converting them, or it can be the user device receiving the signals sent by each end participating in the online conference through a network connection.

[0050] Step S20, obtaining the role volume information of each participant in the audio to be transcribed;

[0051] In this embodiment, specifically, based on preset participant information, the role volume information of each participant is obtained from the audio to be transcribed, wherein the participant is the spokesperson who outputs information in the audio to be transcribed, and the participant information includes voiceprint information, direction information or audio transmission channel information, etc. The participant information can be obtained through a database or set by inputting before the meeting starts, and the role volume information is the volume information of the audio signal corresponding to each participant within a period of time, which can be the average volume value of the audio signal corresponding to each participant in the entire audio to be transcribed, or it can be the average volume value of the audio signal corresponding to each participant in a preset time period in the audio to be transcribed, and can be set according to actual conditions.

[0052] Optionally, the step of obtaining the role volume information of each participant in the audio to be transcribed includes:

[0053] Step A10, decomposing the audio to be transcribed into the role audio of each participant;

[0054] Step A20, respectively obtaining the role volume information corresponding to each participant from the audio of each role.

[0055] In this embodiment, specifically, based on the preset participant information, audio signals having the participant information corresponding to each participant are identified from the audio to be transcribed, the audio signals having the same participant information are synthesized into a role audio, and the role volume information corresponding to each participant is obtained from each role audio, wherein the role audio is the audio signal belonging to the same participant in the audio to be transcribed.

[0056] Optionally, the role volume information includes a segment volume value, and the step of respectively acquiring the role volume information corresponding to each participant from each role audio includes:

[0057] Step A21, taking a preset unit time as a time period, dividing each of the character audios in time sequence to obtain a sub-character audio group in each of the time periods;

[0058] Step A22, obtaining the segment volume value of the sub-role audio of each participant in each of the sub-role audio groups.

[0059] In this embodiment, specifically, a unit time for dividing the role audio is pre-set, and the preset unit time is used as a time period. In terms of timing, the role audio is divided at intervals of the time period to obtain at least one sub-role audio with the same unit time length, and the average volume value of the sub-role audio of each participant in each sub-role audio group is calculated, and the average volume value of the segment is used as the segment volume value of the corresponding sub-role audio segment. It is easy to understand that since the text conversion can be carried out simultaneously with the progress of the meeting, the method of dividing the audio of each role in time is to divide the audio of each role into sub-role audios from morning to night, so that the audio signal to be transcribed that is received first can be calculated and processed first, which can effectively improve the speed of transcribing audio into text, effectively shorten the delay, and make the audio text more synchronized with the audio. The audio of each role is divided into sub-role audios and fragmented calculations and processing can effectively reduce the time spent on the last one-time processing of the larger audio, thereby shortening the delay and making the real-time transcription better.

[0060] Step S30, determining a target participant who theoretically outputs the audio to be transcribed from among all the participants according to the volume information of each role.

[0061] In this embodiment, specifically, the volume information of each of the roles is compared, and according to the proportion of the volume information of each of the roles in the volume information of all the roles, a target participant who theoretically outputs the audio to be transcribed is determined among the participants, wherein the target participant is the person who mainly outputs information in the audio to be transcribed, and the proportion of the volume information of each of the roles in the volume information of all the roles can be the proportion of the volume value of each role in the sum of the volume values ​​of the audio to be transcribed, or it can be that in the audio to be transcribed, in each time period or at each moment, the maximum volume role with the largest volume value is determined, and then the proportion of each participant in the maximum volume role is determined.

[0062] Optionally, the step of determining a target participant who theoretically outputs the audio to be transcribed from among the participants according to the volume information of each role includes:

[0063] Step S31, comparing the volume values ​​of all the segments corresponding to each time period respectively;

[0064] In this embodiment, specifically, according to the time sequence of the time periods, the volume values ​​of all the segments corresponding to each time period are compared in sequence from the earliest to the latest.

[0065] Step S32, determining the participant corresponding to the sub-role audio with the largest segment volume value in each time period as the time period role of the current time period;

[0066] In this embodiment, specifically, the target participant audio segment corresponding to the largest segment volume value among all segment volume values ​​corresponding to each time period is selected, and the participant corresponding to the target participant audio segment is determined as the time period role of this time period, wherein the time period role is the participant who mainly outputs information in the time period. It is easy to understand that the time period role belongs to one of the participants. In an implementable manner, Figure 3 As shown, the time periods are arranged from left to right in order of time from earliest to latest, and the volume values ​​of the fragments of different roles in each time period are obtained, and a comparison is performed within the time period. In the first time period on the left, the volume value of role C is the largest, and role C is used as the time period role of this time period. In the third time period on the left, the volume value of role B is the largest, and role B is used as the time period role of this time period.

[0067] Step S33, determining the time period role with the most repetition times as the target participant of the audio to be transcribed.

[0068] In this embodiment, specifically, the time period roles corresponding to all time periods of the audio to be transcribed are obtained, and the number of repetitions of each time period role is calculated, and the time period role with the largest number of repetitions is determined as the target participant of the audio to be transcribed. In an practicable manner, Figure 3 As shown, if the preset time period is the five time periods from left to right shown in the figure, the time period roles corresponding to the five time periods from left to right are role C, role C, role B, role A and role C, among which role C has the largest number of repetitions, then role C will be used as the target participant of the audio to be transcribed.

[0069] Step S40, transcribing the audio to be transcribed into an audio text having the identity identifier of the target participant.

[0070] In this embodiment, specifically, the audio to be transcribed is transcribed into an initial text, and the initial text is marked according to the determined target participant to generate an audio text with the identity of the target participant, wherein the identity of the target participant is the marking information of the audio text, including the target participant text identifier and the target participant position identifier, etc., so that the user can not only know the text information of the audio to be transcribed through the audio text, but also know who made the sound in the audio to be transcribed. The method of marking the initial text according to the determined target participant can be, on the basis of the initial text, adding an explanatory text of the target participant, for example, adding an explanatory text of "target participant:" before the initial text to indicate that the content after the colon comes from the target participant, or marking the display position information of the initial text as the position corresponding to the target participant, so that when the audio text is displayed, the user can intuitively know that the audio text and the audio to be transcribed come from the target participant.

[0071] In an practicable manner, the step of transcribing the audio to be transcribed into an audio text having the identity identifier of the target participant includes:

[0072] Segment each of the audios to be transcribed according to the time sequence of a preset time period to obtain at least one audio segment to be transcribed;

[0073] Sequentially transcribing the audio segments to be transcribed into audio segment texts according to the time sequence of the time period, and connecting the audio segment texts according to the time sequence of the time period to obtain an initial audio text;

[0074] The initial audio text is marked according to the target participant to obtain an audio text with the identity identifier of the target participant.

[0075] Optionally, after the step of transcribing the audio to be transcribed into an audio text having the identity identifier of the target participant, the method further includes:

[0076] Step B10, generating a time stamp of the audio text according to the reception time information of each of the audio segments to be transcribed;

[0077] Step B20: displaying the audio text according to the time mark via a display device.

[0078] In this embodiment, specifically, the receiving time information of the audio segment to be transcribed is obtained, a time mark of the audio text is generated according to the receiving time information, and the audio text is displayed at the time point corresponding to the time mark through a display device, wherein the time mark is the time information corresponding to the audio text, and the method of generating the time mark of the audio text according to the receiving time information can be to use the receiving time information as the time mark of the audio text, or to use the time information obtained after delaying the receiving time information by a preset delay time as the time mark of the audio text, and the preset delay time can be determined according to actual conditions and actual tests.

[0079] In this embodiment, by generating a time mark and controlling the audio text to be displayed according to the time mark, the synchronous display of the audio to be transcribed and the audio text similar to video subtitles can be achieved. In a conference scenario, the words spoken by different participants can be displayed separately without being mixed together, and the audio text of the current audio content is displayed synchronously. For content that is not clearly heard at the moment, it can be clearly understood through the audio text, thereby effectively improving the meeting experience and meeting effect.

[0080] Optionally, after the step of generating the time mark of the audio text according to the reception time information of each audio segment to be transcribed, the method further includes:

[0081] The audio text and a time stamp of the audio text are saved.

[0082] In this embodiment, the audio text and the time stamp of the audio text are specifically saved locally or in a server as a meeting record so that it can be viewed at any time after the meeting.

[0083] In the present embodiment, by obtaining the audio to be transcribed of the participant's speech between two voice pauses in the conference scene, the role volume information of each participant in the audio to be transcribed is obtained, thereby achieving the acquisition of the role volume information of each participant in the audio to be transcribed with the participation of multiple participants. According to the role volume information, the amount of information output by each participant in the audio to be transcribed can be determined, and then by determining the target participant who theoretically outputs the audio to be transcribed from among the participants according to the role volume information, thereby achieving the determination of the target participant who theoretically outputs the audio to be transcribed in the audio to be transcribed. The transcribed audio is transcribed into an audio text with the identity identification of the target participant, realizing single-channel transcription of the audio to be transcribed with the participation of multiple people, effectively reducing the requirements on the CPU of the user device, and judging the target participant who theoretically outputs the audio to be transcribed in the audio to be transcribed according to the role volume information, overcoming the problem that single-channel transcription cannot distinguish the participants to obtain the text with the participant information, realizing the audio text with accurate participant information through single-channel transcription, and overcoming the technical problem that the transcription of audio with the participation of multiple roles into audio text with participant information has high requirements on CPU performance.

[0084] Further, refer to Figure 4 Based on the above embodiment of the present application, in another embodiment of the present application, the same or similar content as the above embodiment can be referred to the above introduction, and will not be repeated later. On this basis, the step of transcribing the audio to be transcribed into an audio text with the identity of the target participant includes:

[0085] Step S41, synthesizing all sub-role audios corresponding to each time period into an audio package;

[0086] In this embodiment, specifically, all sub-role audios corresponding to each time period are acquired in sequence according to the time sequence of the time periods, and all sub-role audios belonging to the same time period are synthesized into one audio package.

[0087] Step S42, transcribing the audio packets into audio segment texts in chronological order, and connecting the audio segment texts in the order of transcription to obtain an initial audio text;

[0088] In this embodiment, specifically, each of the audio packages is transcribed into audio segment texts in sequence according to the chronological order of the time period to which the audio packages belong, and each of the audio segment texts is connected in the chronological order of the audio segment texts obtained after transcription to obtain the initial audio text.

[0089] Step S43, marking the initial audio text according to the target participant to obtain the audio text with the identity identifier of the target participant.

[0090] In this embodiment, specifically, the initial audio text is marked according to the determined target participant corresponding to the audio to be transcribed, so as to obtain the audio text with the identity identifier of the target participant.

[0091] In this embodiment, by dividing the audio to be transcribed into audio packages according to a time period, and transcribing the audio packages into text in sequence, single-channel transcription can be achieved, and due to the time attribute of the audio, transcription can be started during the transmission of a whole section of audio to be transcribed, without waiting until the whole section of audio to be transcribed is received before transcription is performed, which can effectively improve the overall speed of transcribing audio into text. The division of the time period not only improves the efficiency of transcription, but also can realize the determination of the target participants. The multiplexing of the processing process can effectively reduce the requirements on CPU performance, and achieve single-channel transcription to obtain audio text with accurate participant information, thereby overcoming the technical problem that the transcription of audio involving multiple roles into audio text with participant information has high requirements on CPU performance.

[0092] Furthermore, an embodiment of the present application further provides a single-channel audio-to-text device, which is applied to a single-channel audio-to-text device, and the single-channel audio-to-text device includes:

[0093] A receiving module, used to obtain the audio to be transcribed of the speech of the participant between two voice pauses in the conference scene;

[0094] An acquisition module, used to acquire the role volume information of each participant in the audio to be transcribed;

[0095] A target participant determination module, used to determine a target participant who theoretically outputs the audio to be transcribed from among the participants according to the volume information of each role;

[0096] The transcription module is used to transcribe the audio to be transcribed into an audio text having the identity identifier of the target participant.

[0097] Optionally, the acquisition module is further used to:

[0098] Decomposing the audio to be transcribed into the role audio of each participant;

[0099] The role volume information corresponding to each participant is obtained from each role audio respectively.

[0100] Optionally, the acquisition module is further used to:

[0101] Taking a preset unit time as a time period, dividing each of the role audios in time sequence to obtain a sub-role audio group in each of the time periods;

[0102] Get the segment volume value of the sub-role audio of each participant in each of the sub-role audio groups.

[0103] Optionally, the target participant determination module is further configured to:

[0104] Compare the volume values ​​of all the clips corresponding to each time period respectively;

[0105] The participant corresponding to the sub-role audio with the largest clip volume value in each time period is determined as the time period role of this time period;

[0106] The time period role with the most repetitions is determined as the target participant of the audio to be transcribed.

[0107] Optionally, the transcription module is further used for:

[0108] Combine all sub-role audios corresponding to each time period into one audio package;

[0109] The audio packages are sequentially transcribed into audio segment texts in chronological order, and the audio segment texts are connected in the order of transcription to obtain an initial audio text;

[0110] The initial audio text is marked according to the target participant to obtain an audio text with the identity identifier of the target participant.

[0111] Optionally, the transcription module is further used for:

[0112] Generating a time stamp of the audio text according to the reception time information of each of the audio segments to be transcribed;

[0113] The audio text is displayed according to the time mark by a display device.

[0114] Optionally, the transcription module is further used for:

[0115] The audio text and a time stamp of the audio text are saved.

[0116] The single-channel audio-to-text device provided by the present invention adopts the single-channel audio-to-text method in the above embodiment, which solves the technical problem that the transcription of audio with multiple roles into audio text with participant information requires high CPU performance. Compared with the prior art, the beneficial effects of the single-channel audio-to-text device provided by the embodiment of the present invention are the same as the beneficial effects of the single-channel audio-to-text method provided by the above embodiment, and other technical features in the single-channel audio-to-text device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0117] Furthermore, an embodiment of the present invention provides an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the single-channel audio-to-text method in the above embodiment.

[0118] Reference below Figure 3 , which shows a schematic diagram of the structure of an electronic device suitable for implementing the embodiments of the present disclosure. The electronic device in the embodiments of the present disclosure may include, but is not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 3 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0119] like Figure 3 As shown, the electronic device may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) or a program loaded from a storage device into a random access memory (RAM). In the RAM, various programs and data required for the operation of the electronic device are also stored. The processing device, ROM, and RAM are connected to each other via a bus. An input / output (I / O) interface is also connected to the bus.

[0120] Typically, the following systems can be connected to the I / O interface: input devices including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices including, for example, a magnetic tape, a hard disk, etc.; and communication devices. The communication device can allow the electronic device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows an electronic device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems may be implemented or have alternatively.

[0121] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device, or installed from a ROM. When the computer program is executed by a processing device, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

[0122] The electronic device provided by the present invention adopts the single-channel audio-to-text method in the above embodiment to solve the technical problem that the transcription of audio with multiple roles into audio text with participant information requires high CPU performance. Compared with the prior art, the beneficial effects of the electronic device provided by the embodiment of the present invention are the same as the beneficial effects of the single-channel audio-to-text method provided by the above embodiment, and other technical features in the electronic device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0123] It should be understood that the various parts of the present disclosure can be implemented with hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0124] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art who is familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0125] Furthermore, the present embodiment provides a computer-readable storage medium having computer-readable program instructions stored thereon, and the computer-readable program instructions are used to execute the single-channel audio-to-text conversion method in the above embodiment.

[0126] The computer-readable storage medium provided in the embodiment of the present invention may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination thereof.

[0127] The computer-readable storage medium may be included in the electronic device, or may exist independently without being installed in the electronic device.

[0128] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device: when the vehicle completes parking, obtains parking information and parking space information; calculates at least one actual parking error according to the parking information and the parking space information; determines a result of converting a target single-channel audio into text according to each of the actual parking errors, and outputs prompt information of the result of converting the target single-channel audio into text.

[0129] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0130] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0131] The modules involved in the embodiments described in the present disclosure may be implemented by software or hardware, wherein the name of the module does not limit the unit itself in some cases.

[0132] The computer-readable storage medium provided by the present invention stores computer-readable program instructions for executing the above-mentioned single-channel audio-to-text method, which solves the technical problem that the transcription of audio with multiple roles into audio text with participant information requires high CPU performance. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the embodiment of the present invention are the same as the beneficial effects of the single-channel audio-to-text method provided by the above-mentioned embodiment, and are not described in detail here.

[0133] Furthermore, the present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned single-channel audio-to-text conversion method when executed by a processor.

[0134] The computer program product provided by the present application solves the technical problem that transcribing audio with multiple roles into audio text with participant information requires high CPU performance. Compared with the prior art, the beneficial effects of the computer program product provided by the embodiment of the present invention are the same as the beneficial effects of the single-channel audio-to-text method provided by the above embodiment, which will not be repeated here.

[0135] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent processing scope of the present application.

Claims

1. A method for converting single-channel audio to text, characterized in that: The single-channel audio-to-text method comprises: Obtain the audio to be transcribed of the participant's speech between two voice pauses in the conference scene; Decomposing the audio to be transcribed into the role audio of each participant; Obtaining the role volume information corresponding to each participant from the audio of each role respectively; According to the volume information of each role, determining a target participant from among all the participants to theoretically output the audio to be transcribed; Transcribing the audio to be transcribed into an audio text having the identity identifier of the target participant; The role volume information includes a segment volume value, and the step of obtaining the role volume information corresponding to each participant from each role audio includes: Taking a preset unit time as a time period, dividing each of the role audios in time sequence to obtain a sub-role audio group in each of the time periods; Get the segment volume value of the sub-role audio of each participant in each of the sub-role audio groups.

2. The single-channel audio-to-text method according to claim 1, characterized in that: The step of determining a target participant who theoretically outputs the audio to be transcribed from among the participants according to the volume information of each role includes: Compare the volume values ​​of all the clips corresponding to each time period respectively; The participant corresponding to the sub-role audio with the largest clip volume value in each time period is determined as the time period role of this time period; The time period role with the most repetitions is determined as the target participant of the audio to be transcribed.

3. The single-channel audio-to-text method according to claim 1, characterized in that: The step of transcribing the audio to be transcribed into an audio text having the identity identifier of the target participant includes: Combine all sub-role audios corresponding to each time period into one audio package; The audio packages are sequentially transcribed into audio segment texts in chronological order, and the audio segment texts are connected in the order of transcription to obtain an initial audio text; The initial audio text is marked according to the target participant to obtain an audio text with the identity identifier of the target participant.

4. The single-channel audio-to-text method according to claim 1, characterized in that: After the step of transcribing the audio to be transcribed into an audio text having the identity identifier of the target participant, the method further includes: Generating a time stamp of the audio text according to the reception time information of each of the audio segments to be transcribed; The audio text is displayed according to the time mark by a display device.

5. The single-channel audio-to-text method according to claim 4, characterized in that: After the step of generating the time mark of the audio text according to the receiving time information of each of the audio segments to be transcribed, the method further includes: The audio text and a time stamp of the audio text are saved.

6. A single-channel audio-to-text device, characterized in that: The single-channel audio-to-text device comprises: A receiving module, used to obtain the audio to be transcribed of the speech of the participant between two voice pauses in the conference scene; An acquisition module is used to acquire the role volume information of each participant in the audio to be transcribed; the acquisition module is also used to decompose the audio to be transcribed into the role audio of each participant; respectively acquire the role volume information corresponding to each participant from each role audio; the acquisition module is also used to divide each role audio in time sequence with a preset unit time as a time period to obtain a sub-role audio group in each time period; and acquire the segment volume value of the sub-role audio of each participant in each sub-role audio group; A target participant determination module, used to determine a target participant who theoretically outputs the audio to be transcribed from among the participants according to the volume information of each role; The transcription module is used to transcribe the audio to be transcribed into an audio text having the identity identifier of the target participant.

7. An electronic device, characterized in that: The electronic device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the steps of the single-channel audio-to-text conversion method according to any one of claims 1 to 5.

8. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, on which is stored a program for implementing a method for converting a single-channel audio into text. The program for implementing a method for converting a single-channel audio into text is executed by a processor to implement the steps of the method for converting a single-channel audio into text as claimed in any one of claims 1 to 5.

Images