Audio transcription method, device, electronic device, and storage medium

Abstract

The present invention provides an audio transcription method, device, electronic device, and storage medium. The audio transcription method includes: transcribing audio data into text to obtain an initial transcription text; if the initial transcription text contains an anomaly, determining the anomaly text in the initial transcription text and determining audio segment data corresponding to the anomaly text in the audio data; optimizing the audio segment data and determining new audio data based on the optimized audio segment data; and transcribing the new audio data into text to obtain a target transcription text. This improves the accuracy and efficiency of audio transcription.

Description

Technical Field

[0001] The present invention relates to the field of audio processing technology, and in particular to an audio transcription method, device, electronic device, and storage medium. Background Art

[0002] At present, transcribing audio into text has become a common means to improve work efficiency. Among them, the recognition rate of audio transcription is a key indicator affecting user experience.

[0003] In related audio-to-text technologies, audio recognition can be biased, resulting in low accuracy in audio-to-text transcription. Often, if the transcribed text is inaccurate, users are forced to manually adjust it, reducing audio transcription efficiency. Summary of the Invention

[0004] The present invention aims to solve at least one of the technical problems in the related art to a certain extent. To this end, the present invention provides an audio transcription method, device, electronic device, and storage medium to improve the accuracy and efficiency of audio transcription.

[0005] The present invention provides an audio transcription method, comprising:

[0006] The audio data is transcribed into text to obtain an initial transcribed text; if an abnormality exists in the initial transcribed text, the abnormal text in the initial transcribed text is determined, and audio segment data corresponding to the abnormal text is determined in the audio data; the audio segment data is optimized, and new audio data is determined based on the optimized audio segment data; the new audio data is transcribed into text to obtain a target transcribed text.

[0007] In some feasible embodiments, the optimizing processing of the audio clip data includes: determining the environmental scene to which the audio clip data belongs; performing noise reduction processing on the audio clip data according to the environmental scene and a pre-built environmental scene sound and noise library to obtain optimized audio clip data, wherein the environmental scene sound and noise library includes multiple noise recognition models and the correspondence between each noise recognition model and the environmental scene.

[0008] In some feasible embodiments, the denoising process is performed on the audio clip data according to the environmental scene and a pre-built environmental scene sound and noise library, including: screening out a corresponding noise recognition model from the environmental scene sound and noise library according to the environmental scene; inputting the audio clip data into the screened noise recognition model to determine the noise elements in the audio clip data; and separating the noise elements from the audio clip data to obtain optimized audio clip data.

[0009] In some feasible methods, the steps of constructing the environmental scene sound and noise library include: obtaining training sample sets of different environmental scenes, each training sample set including multiple audio data in the same environmental scene; labeling the training samples in each training sample set; using the labeled training sample set as input and the predicted noise elements as output, training the noise recognition model of the environmental scene corresponding to the training sample set, until all the noise recognition models in the environmental scene sound and noise library are trained, thereby obtaining a constructed environmental scene sound and noise library.

[0010] In some feasible embodiments, after determining the audio segment data corresponding to the abnormal text in the audio data, the method further includes: marking the audio segment data in the audio data; and optimizing the audio segment data includes: optimizing the marked audio segment data.

[0011] In some achievable embodiments, determining new audio data based on the optimized audio segment data includes: replacing corresponding audio segment data in the audio data with the optimized audio segment data to obtain new audio data.

[0012] In some achievable embodiments, after transcribing the audio data into text to obtain an initial transcribed text, the method further includes: determining that an abnormality exists in the initial transcribed text upon detecting receipt of a re-transcription instruction; or determining that an abnormality exists in the initial transcribed text upon detecting that an abnormality exists in the semantics or grammar of the initial transcribed text.

[0013] The present invention also provides an audio transcription device, comprising: a transcription module, used to transcribe audio data into text to obtain an initial transcription text; a determination module, used to determine the abnormal text in the initial transcription text when there is an abnormality in the initial transcription text, and determine the audio segment data corresponding to the abnormal text in the audio data; an optimization module, used to optimize the audio segment data and determine new audio data based on the optimized audio segment data; the transcription module is also used to transcribe the new audio data into text to obtain a target transcription text.

[0014] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the audio transcription method as described above is implemented.

[0015] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the audio transcription methods described above.

[0016] The present invention also provides a computer program product, comprising a computer program, which, when executed by a processor, implements any of the above-described audio transcription methods.

[0017] The audio transcription method, device, electronic device, and storage medium provided by the present invention, when an abnormality exists in the initial transcribed text, determine the abnormal text in the initial transcribed text and determine the audio segment data corresponding to the abnormal text in the audio data; then optimize the audio segment data and determine new audio data based on the optimized audio segment data to improve the clarity of the audio; in this way, the new audio data is then transcribed into text, which can effectively improve the accuracy of the audio transcription, and the abnormal text in the transcribed text can be re-transcribed without the user having to perform manual modification, thereby improving the efficiency of the audio transcription. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 is a schematic structural diagram of an audio transcription system according to an embodiment of the present invention;

[0020] Figure 2 This is one of the flow charts of the audio transcription method provided by an embodiment of the present invention;

[0021] Figure 3 A schematic diagram of the relationship between the initial transcribed text and audio data according to an embodiment of the present invention;

[0022] Figure 4 This is the second flowchart of the audio transcription method provided by an embodiment of the present invention;

[0023] Figure 5 This is the third flowchart of the audio transcription method provided by an embodiment of the present invention;

[0024] Figure 6 This is the fourth flow chart of the audio transcription method provided by an embodiment of the present invention;

[0025] Figure 7 is a block diagram of an audio transcription device according to an embodiment of the present invention;

[0026] Figure 8 It is a structural block diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0027] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0028] Figure 1 2 is a schematic diagram of the structure of an audio transcription system according to an embodiment of the present invention.

[0029] like Figure 1 As shown, the system includes an audio collector, an audio transcription device, an audio processing device, and an audio optimization device. The input end of the audio transcription device is connected to the audio collector to receive audio data sent by the audio collector; the output end of the audio transcription device can be connected to the display device of the electronic device. The audio transcription device can send the text data obtained by audio transcription to the display device, and the display device displays the transcribed text. The audio transcription device also interacts with the audio processing device, which in turn interacts with the audio beautification device.

[0030] For ease of understanding, the audio transcription method of an embodiment of the present invention is introduced below in conjunction with the above-mentioned audio transcription system.

[0031] Figure 2 This is one of the flow charts of the audio transcription method provided by the embodiment of the present invention. Figure 2 As shown, the audio transcription method may include the following steps:

[0032] Step 210: transcribe the audio data into text to obtain an initial transcribed text.

[0033] In this embodiment, the audio data may be audio collected in real time or pre-recorded audio.

[0034] As an example, an audio collector can be used to collect audio data. Specifically, the audio collector can collect surrounding sound signals and convert the sound signals into audio data. The data format of the audio data can be a standard WAV data format. In addition, the audio collector can be provided with a parameter adjustment function. For example, parameters such as the sampling rate, the number of sampling bits, and the number of channels can be adjusted. By adjusting the parameters such as the sampling rate, the number of sampling bits, and the number of channels of the audio collector, the accuracy of the sound signal collected by the audio collector can be adjusted.

[0035] After the audio collector obtains the audio data, the audio collector will transmit the audio data to the audio transcription device, which will then extract the voice audio in the audio data, recognize and transcribe the voice audio to obtain the initial transcription text.

[0036] In some embodiments, a VAD (Voice Activity Detection) detection method can be used to detect the voice endpoint VAD in the audio data. After determining the voice endpoint VAD in the audio data, the audio data within the coverage range of the voice endpoint VAD is extracted to obtain the voice audio, thereby realizing the extraction of the voice audio in the audio data.

[0037] In some embodiments, the output end of the audio transcription device can be connected to a display device of an electronic device, and the initial transcription text obtained by transcription can be displayed through the display device so that the user can view the audio transcription result.

[0038] Step 220: When an abnormality exists in the initial transcribed text, the abnormal text in the initial transcribed text is determined, and audio segment data corresponding to the abnormal text is determined in the audio data.

[0039] It is understandable that the audio data not only contains the voice audio emitted by the user, but also environmental sounds, background sounds, and noises, etc., and these noises will affect the transcription recognition rate of the audio data. Therefore, there may be transcription errors in the initial transcription text obtained by transcription.

[0040] In this embodiment, if it is determined that the initial transcribed text contains transcription errors, the erroneously transcribed text needs to be re-transcribed. Specifically, the audio transcription device sends the initial transcribed text to the audio processing device, and the audio processing device determines abnormal text in the initial transcribed text. It should be noted that the abnormal text can be a sentence or word that is transcribed incorrectly in the initial transcribed text.

[0041] In some embodiments, the methods for determining abnormal text in the initial transcription include (but are not limited to) the following two methods. One method is that the user can manually select abnormal text in the initial transcription. As an example, the display screen for displaying the initial transcription can be a touch screen, and the user can select abnormal text in the initial transcription through touch operation. The other method is to detect whether there are semantic or grammatical errors in the initial transcription, and the text with semantic or grammatical errors can be regarded as abnormal text.

[0042] After determining the abnormal text in the initial transcription, the audio processing device will determine the audio segment data corresponding to the abnormal text in the audio data. It is understandable that a section of text in the initial transcription may correspond to multiple voice endpoint VADs in the audio data. By performing VAD detection on the audio data, the voice endpoint VAD contained in each section of text in the initial transcription can be obtained. Based on the voice endpoint VAD, the audio start time and audio end time of the text section can be determined, and then the correspondence between the initial transcription and the audio data can be obtained.

[0043] Figure 3 Schematic diagram of the relationship between the initial transcribed text and audio data in an embodiment of the present invention. As an example, Figure 3 As described above, text1 in the initial transcription text contains 1-n voice endpoints VAD. According to the position information of the voice endpoints VAD in the audio data, it can be determined that the audio segment data corresponding to text1 are audio segment 1 and audio segment 2.

[0044] Therefore, the above method can be used to determine the audio segment data corresponding to the abnormal text in the audio data.

[0045] Step 230: Optimize the audio segment data and determine new audio data based on the optimized audio segment data.

[0046] After the audio processing device determines the audio segment data corresponding to the abnormal text, the audio processing device will send the audio segment data to the audio optimization device, and the audio optimization device will optimize the audio segment data to make this part of the audio segment data clearer, thereby improving the recognition rate of this part of the audio segment data.

[0047] It should be noted that the optimization processing of the audio segment data may include noise reduction processing, sound enhancement processing, sound beautification processing, etc. of the audio segment data, which is not specifically limited here.

[0048] After the audio optimization device optimizes the audio segment data, the audio optimization device sends the optimized audio segment to the audio processing device; the audio processing device reintegrates the optimized audio segment into the original audio data to obtain new audio data.

[0049] Step 240: perform text transcription on the new audio data to obtain a target transcribed text.

[0050] After receiving the new audio data, the audio processing device sends the new audio data to the audio transcription device. The audio transcription device then re-transcribes the new audio data to obtain the target transcription text. After obtaining the target transcription text, the audio transcription device can send the target transcription text to the display device of the electronic device, so that the display device can display the target transcription text so that the user can view the audio transcription results.

[0051] Figure 4 This is the second flow chart of the audio transcription method provided by the embodiment of the present invention. Figure 4 As shown, the following steps can be used to optimize the audio segment data:

[0052] Step 410: Determine the environment scene to which the audio segment data belongs.

[0053] Step 420: performing noise reduction processing on the audio segment data according to the environmental scene and the pre-built environmental scene sound and noise library to obtain optimized audio segment data.

[0054] It should be noted that the environmental scene sound noise library includes noise recognition models for multiple different environmental scenes, as well as a correspondence between each noise recognition model and the environmental scene.

[0055] Specifically, after receiving the audio segment data, the audio optimization device can first determine the environmental scene to which the audio segment data belongs. The environmental scene of the audio segment data can be a meeting scene, an outdoor scene, a classroom scene, etc.

[0056] The method for determining the environmental scene to which the audio clip data belongs may be: the user manually selects the environmental scene. As an example, a user interface may be displayed on a display device of the electronic device, wherein a plurality of virtual icons representing different environmental scenes are displayed in the user interface, and the user may select the environmental scene to which the audio clip data belongs by clicking on the virtual icon.

[0057] Another method for determining the environmental scene to which the audio clip data belongs is: the audio optimization device can be provided with an audio scene recognition function to automatically identify the environmental scene based on the audio clip data. The audio scene recognition function can be implemented using an existing audio scene recognition model, which is not described in detail here. After the audio optimization device identifies the environmental scene using the audio scene recognition function, the user can also manually calibrate the environmental scene to ensure the accuracy of the environmental scene recognition to which the audio clip data belongs.

[0058] After determining the environmental scene to which the audio clip data belongs, a corresponding noise recognition model can be matched for the audio clip data in a pre-built environmental scene sound and noise library according to the environmental scene, and the audio clip data can be input into the noise recognition model, and the noise in the audio clip can be identified using the noise recognition model.

[0059] Figure 5 This is the third flow chart of the audio transcription method provided by the embodiment of the present invention. Figure 5 As shown, the further optimization processing steps for the audio segment data include:

[0060] Step 510: Filter out a corresponding noise recognition model from an environmental scene sound and noise library according to the environmental scene.

[0061] Step 520: Input the audio segment data into the selected noise recognition model to determine the noise elements in the audio segment data.

[0062] Step 530: Separate the noise element from the audio segment data to obtain optimized audio segment data.

[0063] Specifically, the audio optimization device selects a corresponding noise recognition model from the environmental scene sound and noise library according to the environmental scene to which the audio clip data belongs. The noise recognition model is trained using an audio data set of the same environmental scene. Therefore, the noise recognition model can recognize the ambient sound, background sound, and noise in the audio data under the environmental scene.

[0064] After selecting the noise recognition model to use, the audio optimization device inputs the audio segment data into the noise recognition model, which then outputs the noise elements in the audio segment data. The audio optimization device then separates these noise elements from the audio segment data, thereby achieving noise reduction on the audio segment data and obtaining optimized audio segment data.

[0065] As an example, the environmental scene sound noise library contains multiple noise recognition models, including noise recognition models for conference scenes, noise recognition models for outdoor scenes, and noise recognition models for classroom scenes. The audio optimization device automatically recognizes that the environmental scene to which the audio clip data belongs is a conference scene through the audio scene recognition function, and the audio optimization device inputs the audio clip data into the noise recognition model of the conference scene. The noise recognition model of the conference scene will identify the noise elements in the audio clip data, such as keyboard sounds, applause, background music, etc. The audio optimization device separates the noise elements in the audio clip data, thereby highlighting the human voice in the audio clip data and improving the clarity of the speech audio in the audio clip data.

[0066] In some embodiments, the environmental scene sound and noise library can be constructed by: obtaining training sample sets of different environmental scenes, each training sample set including multiple audio data in the same environmental scene; labeling the training samples in each training sample set; using the labeled training sample set as input and the predicted noise elements as output, training the noise recognition model of the environmental scene corresponding to the training sample set until all noise recognition models in the environmental scene sound and noise library are trained, thereby obtaining a constructed environmental scene sound and noise library.

[0067] It should be understood that the environmental scene sound noise library in this embodiment contains noise recognition models for multiple environmental scenes, and each noise recognition model is used to identify noise elements in audio data from only one environmental scene. Therefore, noise recognition models for different environmental scenes require training sample sets from different environmental scenes for training.

[0068] The training sample set includes multiple audio data in the same environmental scene. For example, multiple groups of conference audio data can be collected, and these conference audio data are used as training sample sets for training the conference scene noise recognition model.

[0069] After obtaining the training sample set, the training samples in the training sample set need to be labeled. The labeling method can adopt the existing labeling method and will not be described in detail here. During model training, the labeled training samples are input into the noise recognition model, and the noise recognition model outputs the predicted noise elements, thereby training the noise recognition model. The other noise recognition models in the environmental scene sound noise library are trained in the same way until all noise recognition models in the environmental scene sound noise library are trained. The constructed environmental scene sound noise library can be obtained.

[0070] In some embodiments, after determining the audio segment data corresponding to the abnormal text in the audio data, the audio transcription method further includes: marking the audio segment data in the audio data. In step 230, the audio segment data is optimized, including: optimizing the marked audio segment data.

[0071] Specifically, the audio processing device can be provided with a marking function. After determining the audio segment data corresponding to the abnormal text, the audio processing device can mark the audio segment data corresponding to the abnormal text. After marking the audio segment data, the audio processing device segments the marked audio segment data and sends the marked audio segment data to the audio optimization device, which then optimizes the marked audio segment data.

[0072] It should be noted that the mark of the audio segment data will be stored in the audio data, so that the position of the audio segment data can be determined later according to the mark.

[0073] like Figure 3 As shown in the example, six audio segments are divided from the audio data. The audio processing device determines, based on the abnormal text, that the audio segments to be optimized are: audio segment data 1, audio segment data 2, and audio segment data 5. The audio processing device then marks these three audio segments. After marking, the audio processing device sends the marked audio segment data 1, audio segment data 2, and audio segment data 5 to the audio optimization device.

[0074] In some implementations, a method of determining new audio data based on the optimized audio segment data may be: replacing corresponding audio segment data in the audio data with the optimized audio segment data to obtain new audio data.

[0075] Specifically, after obtaining the optimized audio segment data, the audio optimization device sends the optimized audio segment data to the audio processing device. The audio processing device then replaces the audio segment data at the marked position in the original audio data with the optimized audio segment data. After the replacement of each optimized audio segment data is completed, a new audio data is obtained.

[0076] In some embodiments, after step 210, the audio transcription method further includes: determining that an abnormality exists in the initial transcription text when a re-transcription instruction is detected; or determining that an abnormality exists in the initial transcription text when an abnormality exists in the semantics or grammar of the initial transcription text.

[0077] Specifically, after obtaining the initial transcription text, it is necessary to determine whether the initial transcription text has an abnormality. If the initial transcription text has an abnormality, the process proceeds to step 220. In this embodiment, determining whether the initial transcription text has an abnormality can be done in either of the following two ways.

[0078] A method for allowing a user to view an initial transcribed text and determine whether there is an abnormality in the initial transcribed text. Specifically, after the audio transcription device completes the transcription of the audio data, the audio transcription device sends the initial transcribed text to the display device of the electronic device, and the display device displays the initial transcribed text to the user. A re-transcription control can also be set in the user interface displayed by the display device. The user checks the displayed initial transcribed text to confirm whether there is an abnormality in the initial transcribed text. If the user believes that there is an abnormality in the initial transcribed text, the user can click the re-transcription control through a touch operation. When it is detected that a re-transcription instruction is received, it is determined that there is an abnormality in the initial transcribed text.

[0079] Another method is to automatically determine whether the initial transcription is abnormal by determining whether there are semantic or grammatical errors in the initial transcription. If there are semantic or grammatical errors in the initial transcription, it is determined that the initial transcription is abnormal.

[0080] For ease of understanding, the audio transcription method according to an embodiment of the present invention is introduced below through a specific example. Figure 6 This is the fourth flow chart of the audio transcription method provided by the embodiment of the present invention. Figure 6 As shown, the audio transcription method may further include the following steps:

[0081] Step 610: Acquire audio data.

[0082] Step 620: transcribe the audio data into text to obtain an initial transcribed text.

[0083] Step 630: Determine whether the initial transcription text has any anomaly. If the initial transcription text has any anomaly, proceed to step 640; if the initial transcription text does not have any anomaly, output the initial transcription text.

[0084] Step 640: Determine abnormal text in the initial transcription text.

[0085] Step 650: Determine the audio segment data corresponding to the abnormal text in the audio data, and mark the audio segment data.

[0086] Step 660: The marked audio segment data is subjected to noise reduction processing, sound enhancement processing, and sound beautification processing to obtain optimized audio segment data.

[0087] Step 670: Replace the corresponding audio segment data in the audio data with the optimized audio segment data to obtain new audio data.

[0088] Step 680: perform text transcription on the new audio data to obtain a target transcribed text.

[0089] Thus, when an abnormality exists in the initial transcribed text obtained by audio transcription, the abnormal text is found in the initial transcribed text, and the audio segment data corresponding to the abnormal text in the audio data is determined; the audio segment data is then optimized; and the optimized audio segment data replaces the original audio segment data to obtain new audio data. Through the optimization process, the new audio data is clearer than the original audio data, and thus can be more accurately identified and transcribed. Therefore, the target transcribed text obtained by re-transcribing the new audio data is also more accurate, thereby improving the accuracy of the audio transcription. In addition, when an abnormality exists in the initial transcribed text obtained by transcription, the abnormal text in the transcribed text can be re-transcribed without the user having to manually modify it, thereby improving the efficiency of audio transcription.

[0090] Corresponding to the above embodiment, the embodiments of this specification also propose an audio transcription device. Figure 7 4 is a block diagram of an audio transcription device according to an embodiment of the present invention.

[0091] like Figure 7 As shown, the device 700 includes a transcription module 710 for transcribing audio data into text to obtain an initial transcription text; a determination module 720 for determining the abnormal text in the initial transcription text when the initial transcription text contains an abnormality, and determining the audio segment data corresponding to the abnormal text in the audio data; and an optimization module 730 for optimizing the audio segment data and determining new audio data based on the optimized audio segment data. The transcription module is also used to transcribe the new audio data into text to obtain a target transcription text.

[0092] Thus, the audio data is transcribed into text by the transcription module 710 to obtain a transcription of the initial transcription text. When there is an abnormality in the initial transcription text, the abnormal text in the initial transcription text is determined by the determination module 720, and the audio segment data corresponding to the abnormal text is determined in the audio data. The audio segment data is optimized by the optimization module 730, and new audio data is determined based on the optimized audio segment data to improve the clarity of the audio; in this way, the new audio data is transcribed into text, which can effectively improve the accuracy of the audio transcription, and the abnormal text in the transcribed text can be re-transcribed without the user making manual modifications, thereby improving the efficiency of the audio transcription.

[0093] It should be noted that for details not disclosed in the audio transcription device of this embodiment, please refer to the details disclosed in the embodiment of the audio transcription method in the embodiments of this specification, and no further details will be given here.

[0094] Figure 8 An example of a physical structure diagram of an electronic device is shown below. Figure 8As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830, and a communication bus 840, wherein the processor 810, the communication interface 820, and the memory 830 communicate with each other via the communication bus 840. The processor 810 may call the logic instructions in the memory 830 to execute the audio transcription method, which includes: performing text transcription on the audio data to obtain an initial transcription text; if there is an abnormality in the initial transcription text, determining the abnormal text in the initial transcription text and determining audio segment data corresponding to the abnormal text in the audio data; optimizing the audio segment data and determining new audio data based on the optimized audio segment data; and performing text transcription on the new audio data to obtain a target transcription text.

[0095] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0096] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the audio transcription method provided by the above methods, which includes: transcribing audio data into text to obtain an initial transcribed text; when there is an abnormality in the initial transcribed text, determining the abnormal text in the initial transcribed text, and determining audio segment data corresponding to the abnormal text in the audio data; optimizing the audio segment data, and determining new audio data based on the optimized audio segment data; and transcribing the new audio data into text to obtain a target transcribed text.

[0097] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the audio transcription method provided by the above-mentioned methods, the method comprising: transcribing audio data into text to obtain an initial transcribed text; in the event that there is an abnormality in the initial transcribed text, determining the abnormal text in the initial transcribed text, and determining audio segment data corresponding to the abnormal text in the audio data; optimizing the audio segment data, and determining new audio data based on the optimized audio segment data; and transcribing the new audio data into text to obtain a target transcribed text.

[0098] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0099] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiments.

[0100] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. An audio transcription method, characterized in that: include: Transcribe the audio data into text to obtain an initial transcribed text; If an abnormality exists in the initial transcribed text, determining the abnormal text in the initial transcribed text, and determining audio segment data corresponding to the abnormal text in the audio data; Optimizing the audio segment data and determining new audio data based on the optimized audio segment data; Performing text transcription on the new audio data to obtain a target transcribed text; The optimizing process of the audio segment data includes: Determining the environmental scene to which the audio clip data belongs; According to the environmental scene and a pre-built environmental scene sound and noise library, the audio clip data is subjected to noise reduction processing to obtain optimized audio clip data, wherein the environmental scene sound and noise library includes multiple noise recognition models and a corresponding relationship between each noise recognition model and the environmental scene.

2. The audio transcription method according to claim 1, wherein: The performing noise reduction processing on the audio segment data according to the environmental scene and a pre-built environmental scene sound and noise library includes: Filtering a corresponding noise recognition model from the environmental scene sound noise library according to the environmental scene; Inputting the audio segment data into the selected noise recognition model to determine noise elements in the audio segment data; The noise element is separated from the audio segment data to obtain optimized audio segment data.

3. The audio transcription method according to claim 2, wherein: The steps of constructing the environmental scene sound noise library include: Acquire training sample sets of different environmental scenes, each of the training sample sets including a plurality of audio data in the same environmental scene; Labeling each training sample in the training sample set; Taking the labeled training sample set as input and the predicted noise elements as output, the noise recognition model of the environmental scene corresponding to the training sample set is trained until all the noise recognition models in the environmental scene sound noise library are trained to obtain the environmental scene sound noise library.

4. The audio transcription method according to any one of claims 1 to 3, characterized in that: After determining the audio segment data corresponding to the abnormal text in the audio data, the method further includes: Marking the audio segment data in the audio data; The optimizing process of the audio segment data includes: The marked audio segment data is optimized.

5. The audio transcription method according to any one of claims 1 to 3, characterized in that: The determining of new audio data based on the optimized audio segment data includes: The optimized audio segment data is used to replace corresponding audio segment data in the audio data to obtain new audio data.

6. The audio transcription method according to any one of claims 1 to 3, characterized in that: After transcribing the audio data into text to obtain an initial transcribed text, the method further includes: In the case of detecting that a re-transcription instruction is received, determining that an abnormality exists in the initial transcription text; or When there is an anomaly in the semantics or grammar of the initial transcribed text, it is determined that there is an anomaly in the initial transcribed text.

7. An audio transcription device, characterized in that: include: A transcription module is used to transcribe the audio data into text to obtain an initial transcription text; a determination module, configured to, when an abnormality exists in the initial transcribed text, determine the abnormal text in the initial transcribed text, and determine audio segment data corresponding to the abnormal text in the audio data; an optimization module, configured to optimize the audio segment data and determine new audio data based on the optimized audio segment data; The transcription module is further configured to perform text transcription on the new audio data to obtain a target transcribed text; The optimization module is specifically used for: Determining the environmental scene to which the audio clip data belongs; According to the environmental scene and a pre-built environmental scene sound and noise library, the audio clip data is subjected to noise reduction processing to obtain optimized audio clip data, wherein the environmental scene sound and noise library includes multiple noise recognition models and a corresponding relationship between each noise recognition model and the environmental scene.

8. An electronic device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the program, the audio transcription method according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the audio transcription method according to any one of claims 1 to 6 is implemented.

Images