Delayed transmission method, device, terminal and storage medium for audio data

Abstract

The present invention provides a delayed transmission method, device, terminal and storage medium for audio data. The delayed transmission method for audio data includes: obtaining the same audio information transmitted by at least two audio channels respectively; the audio information includes voice data; performing voice recognition on the audio information transmitted by the at least two audio channels respectively, and determining the recognition time of the voice data; and determining the number of queue-cached audio frames corresponding to the at least two audio channels respectively based on the recognition time corresponding to the at least two audio channels. The present application identifies the audio information transmitted by different audio channels, and determines the number of queue-cached audio frames of the audio channels based on the recognition time corresponding to all audio channels respectively, so that the mixing module can receive audio data transmitted synchronously by different audio channels, thereby improving the mixing effect.

Description

Technical Field

[0001] The present invention relates to the technical field of audio data processing, and in particular to a delayed transmission method, device, terminal and storage medium for audio data. Background Art

[0002] Voice mixing is a crucial component of multimedia conferencing. Because the mixed audio sources originate from different devices and travel through different transmission paths, there's a delay between the time each channel's captured audio data actually reaches the mixing module and the time the real-world sound is generated. This delay can vary significantly between channels, especially for audio data transmitted over a network. Because it undergoes additional processing steps, such as encoding, network transmission, and decoding, compared to analog audio capture, its latency can be significantly higher than that of analog audio sources. Different audio capture devices can also exhibit varying latency due to internal processing flows and network fluctuations. If these delay differences are not addressed and the audio data is fed directly into the mixing module, the mixed audio data may contain overlapping sounds, severely impacting the quality of the mix. Summary of the Invention

[0003] The main technical problem solved by the present invention is to provide a delayed transmission method, device, terminal and storage medium for audio data, so as to solve the problem in the prior art that the time of audio source data of various channels is not synchronized, thereby affecting the mixing effect.

[0004] To solve the above technical problems, the first technical solution adopted by the present invention is: to provide a delayed transmission method for audio data, the delayed transmission method for audio data including: obtaining the same audio information transmitted by at least two audio channels respectively; the audio information includes voice data; performing voice recognition on the audio information transmitted through at least two audio channels respectively, and determining the recognition time of the voice data; based on the recognition time corresponding to the at least two audio channels respectively, determining the number of queue cache audio frames corresponding to the at least two audio channels respectively.

[0005] Among them, voice recognition is performed on audio information transmitted through at least two audio channels respectively to determine the recognition time of the voice data, including: performing voice recognition on the audio information; judging whether the audio information recognizes text information of the voice data; in response to the text information of the voice data being recognized, determining the recognition time of the recognized text information; wherein the recognition time is the moment when the text information corresponding to the voice data is recognized.

[0006] In which, the text information includes at least one word text; in response to the text information of the recognized voice data, the recognition time of the recognized text information is determined, including: saving the recognized word text and the recognition time of the recognized word text in the data queue in sequence; in response to the queue length of the data queue reaching a preset length, the recognition time of the last word text saved in the data queue is determined as the recognition time of the text information.

[0007] Among them, based on the recognition time corresponding to at least two audio channels, the number of queue cache audio frames corresponding to at least two audio channels is determined, including: judging whether the text information stored in the data queue of the preset length corresponding to each audio channel is the same; in response to the text information stored in the data queue of the preset length corresponding to each audio channel being the same, the number of queue cache audio frames corresponding to each audio channel is determined according to the recognition time of the same word text in the text information corresponding to each audio channel.

[0008] Among them, based on the recognition times corresponding to at least two audio channels, the number of queue cache audio frames corresponding to at least two audio channels is determined, including: selecting the latest recognition time among the recognition times corresponding to at least two audio channels; and calculating the difference between the recognition time corresponding to each audio channel and the latest recognition time to determine the number of queue cache audio frames corresponding to each audio channel.

[0009] Among them, the difference between the recognition time corresponding to each audio channel and the latest recognition time is calculated respectively to determine the number of queue cache audio frames corresponding to each audio channel, including: calculating the difference between the recognition time corresponding to the audio channel and the latest recognition time; determining the number of queue cache audio frames of the audio channel according to the difference, the sampling rate of the audio information and the number of audio samples of each audio frame in the audio information.

[0010] Among them, based on the identification time corresponding to at least two audio channels, the number of queue cache audio frames corresponding to at least two audio channels is determined, and then it includes: adding preset audio frames of the corresponding queue cache audio frames to the queue of each audio channel.

[0011] To solve the above technical problems, the second technical solution adopted by the present invention is: to provide a delayed transmission device for audio data, the delayed transmission device for audio data including: an acquisition module for acquiring the same audio information transmitted by at least two audio channels respectively; the audio information includes voice data; a recognition module for performing voice recognition on the audio information transmitted through at least two audio channels respectively, and determining the recognition time of the voice data; a processing module for determining the number of queue cache audio frames corresponding to at least two audio channels respectively based on the recognition times corresponding to at least two audio channels respectively.

[0012] In order to solve the above technical problems, the third technical solution adopted by the present invention is: to provide a terminal, which includes a memory, a processor, and a computer program stored in the memory and running on the processor, and the processor is used to execute program data to implement the steps in the above-mentioned delayed transmission method of audio data.

[0013] In order to solve the above technical problems, the fourth technical solution adopted by the present invention is: providing a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps in the above-mentioned delayed transmission method of audio data are implemented.

[0014] The beneficial effects of the present invention are: different from the prior art, a delayed transmission method, device, terminal and storage medium for audio data are provided, and the delayed transmission method for audio data includes: obtaining the same audio information transmitted by at least two audio channels respectively; the audio information includes voice data; performing voice recognition on the audio information transmitted by at least two audio channels respectively, and determining the recognition time of the voice data; based on the recognition time corresponding to the at least two audio channels respectively, determining the number of queue cached audio frames corresponding to the at least two audio channels respectively. The present application identifies the audio information transmitted by different audio channels, and determines the number of queue cached audio frames of the audio channels based on the recognition time corresponding to all audio channels respectively, so that the mixing module can receive audio data transmitted synchronously by different audio channels, thereby improving the mixing effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0016] Figure 1 1 is a flow chart of the method for delayed transmission of audio data provided by the present invention;

[0017] Figure 2 It is a flowchart of a specific embodiment of the method for delayed transmission of audio data provided by the present invention;

[0018] Figure 3 It is the storage structure of the data queue FIFO provided by the present invention;

[0019] Figure 4 is a schematic block diagram of the delayed transmission device for audio data provided by the present invention;

[0020] Figure 5is a schematic block diagram of an embodiment of a terminal provided by the present invention;

[0021] Figure 6 It is a schematic block diagram of an embodiment of a computer-readable storage medium provided by the present invention. DETAILED DESCRIPTION

[0022] The following describes the embodiments of the present application in detail with reference to the accompanying drawings.

[0023] In the following description, for the purpose of explanation rather than limitation, specific details such as specific system structures, interfaces, and technologies are provided to facilitate a thorough understanding of the present application.

[0024] The term "and / or" in this document simply describes a relationship between related objects, indicating that three possible relationships exist. For example, "A and / or B" can mean: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " in this document generally indicates that the related objects are in an "or" relationship. Furthermore, "many" in this document means two or more than two.

[0025] In order to enable those skilled in the art to better understand the technical solution of the present invention, a delayed transmission method for audio data provided by the present invention is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0026] See also Figure 1 , Figure 1 : It is a flow chart of the delayed transmission method of audio data provided by the present invention. In this embodiment, a delayed transmission method of audio data is provided, which can be applied to recording and broadcasting products in the education industry, and can also be applied to industries such as conference recording and broadcasting products. The audio data collected by each audio channel is transmitted to the recording and broadcasting host for mixing. Prior to this, each audio channel needs to be calibrated to ensure that each audio channel can achieve synchronization before mixing. The delayed transmission method of audio data includes the following steps.

[0027] S11: Acquire the same audio information transmitted by at least two audio channels respectively.

[0028] Specifically, each audio channel transmits the collected audio data to the recording and broadcasting host, and the audio information can be transmitted to the recording and broadcasting host by wireless transmission and / or wired transmission. Among them, the transmission method of the audio channel transmitting the audio data to the mixing module in the recording and broadcasting host includes wired transmission, Bluetooth transmission, WIFI transmission, Zigbee transmission and other transmission methods. Since different audio data can transmit the collected audio data containing voice data to the mixing module in the recording and broadcasting host through different transmission methods, the transmission speeds of different transmission methods are different, and in the standard synchronous sound source transmission process, the methods of encoding and decoding voice data are also different, which makes the time for different audio channels to transmit to the mixing module delayed, so that different audio channels cannot synchronously transmit the collected voice data to the mixing module. Among them, the audio information includes voice data.

[0029] S12: Performing voice recognition on the audio information transmitted through at least two audio channels respectively, and determining the recognition time of the voice data.

[0030] Specifically, voice recognition is performed on the audio information; it is determined whether the audio information recognizes the text information of the voice data; in response to the text information of the voice data being recognized, the recognition time of the recognized text information is determined; wherein the recognition time is the moment when the text information corresponding to the voice data is recognized. In a specific embodiment, the text information includes at least one word text; the recognized word text and the recognition time of the recognized word text are sequentially saved to the data queue; in response to the queue length of the data queue reaching a preset length, the recognition time of the last word text saved in the data queue is determined as the recognition time of the text information. wherein, the recognition time of the voice data refers to the moment when the voice data is recognized. Specifically, the recognition time of the voice data can be the moment when a preset word or preset sentence in the voice data is recognized.

[0031] S13: Determine the number of audio frames buffered in the queues corresponding to the at least two audio channels, based on the recognition times corresponding to the at least two audio channels.

[0032] Specifically, it is determined whether the text information stored in the data queues of preset length corresponding to each audio channel is the same; in response to the text information corresponding to each audio channel being the same, the number of queue cache audio frames corresponding to at least two audio channels is determined based on the recognition time of the same word text in the text information corresponding to each audio channel.

[0033] In a specific embodiment, the latest recognition time among the recognition times corresponding to at least two audio channels is selected; and the difference between the recognition time corresponding to each audio channel and the latest recognition time is calculated to determine the number of audio frames buffered in the queue corresponding to each audio channel.

[0034] Specifically, the difference between the recognition time corresponding to the audio channel and the latest recognition time is calculated; and the number of audio frames in the queue cache of the audio channel is determined based on the difference, the sampling rate of the audio information, and the number of audio samples of each audio frame in the audio information.

[0035] In another embodiment, a preset number of audio frames corresponding to the queue buffer audio frames is added to the queue of each audio channel.

[0036] The present embodiment provides a delayed transmission method for audio data, including: obtaining the same audio information transmitted by at least two audio channels respectively; the audio information includes voice data; performing voice recognition on the audio information transmitted by at least two audio channels respectively, and determining the recognition time of the voice data; and determining the number of queue-cached audio frames corresponding to the at least two audio channels respectively based on the recognition time corresponding to the at least two audio channels. The present application identifies the audio information transmitted by different audio channels, and determines the number of queue-cached audio frames of the audio channels based on the recognition time corresponding to all audio channels respectively, so that the mixing module can receive audio data transmitted synchronously by different audio channels, thereby improving the mixing effect.

[0037] See also Figure 2 , Figure 2 1 is a flow chart of a specific embodiment of the delayed transmission method of audio data provided by the present invention. This embodiment provides a delayed transmission method of audio data, which can be applied to recording and broadcasting products in the education industry, and can also be applied to industries such as conference recording and broadcasting products. The audio data collected by each audio channel is transmitted to the recording and broadcasting host for mixing. Prior to this, each audio channel needs to be calibrated to ensure that each audio channel can achieve synchronization before mixing. The delayed transmission method of audio data includes the following steps.

[0038] S201: At least two audio channels simultaneously collect the same voice data.

[0039] Specifically, the voice data corresponding to the speech of a natural person is collected by at least two audio devices, that is, the voice data is collected simultaneously by at least two audio channels. Among them, the audio channels can be two or more. Among them, the audio channel can be a microphone, an analog high-definition camera, an IP camera (IP Camera, IPC) and / or a network dome camera, or other audio acquisition devices, which are not limited here. The voice data can be the voice information spoken by a natural person, and the voice information can be the speech of a natural person. When outputting the voice information, the natural person needs to keep the environment quiet and speak at a relatively high volume, so as to facilitate the subsequent recognition of the collected audio information. In one embodiment, the audio frames stored in each audio channel are initialized before the voice data is collected. In other words, the number of audio frames stored in each audio channel is zero.

[0040] S202: Acquire the same audio information transmitted by at least two audio channels respectively.

[0041] Specifically, each audio channel transmits the collected audio data to the recording and broadcasting host, and can transmit the audio information to the recording and broadcasting host by wireless transmission and / or wired transmission. Among them, the transmission method of the audio channel transmitting the audio data to the mixing module in the recording and broadcasting host includes wired transmission, Bluetooth transmission, WIFI transmission, Zigbee transmission and other transmission methods. Since different audio data can transmit the collected audio data containing voice data to the mixing module in the recording and broadcasting host by different transmission methods, the transmission speeds of different transmission methods are different, and in the process of standard synchronous sound source transmission, the methods of encoding and decoding voice data are also different, which makes the time for different audio channels to transmit to the mixing module delayed, so that different audio channels cannot synchronously transmit the collected voice data to the mixing module.

[0042] S203: Perform voice recognition on the audio information.

[0043] Specifically, the received audio information is directly subjected to speech recognition by the speech recognition module. When the audio information contains audio data with corresponding text, the speech recognition module can identify the text information corresponding to the audio information. If the audio information is audio data without text, the speech recognition module cannot identify the corresponding text information.

[0044] In another optional embodiment, the received audio information can also be subjected to speech recognition to obtain corresponding text information. In order to more accurately determine that the recognized text information is the text information of the voice data in the audio information, it is necessary to confirm the recognized text information again.

[0045] S204: Determine whether the audio information recognizes text information of the voice data.

[0046] Specifically, the audio data is recognized by the voice recognition module. If the audio information is recognized to contain voice data, the process directly jumps to step S205. If the voice data in the audio data is not recognized, it indicates that the voice data has not been collected, and the process directly jumps to step S201.

[0047] S205: Determine the recognition time of the recognized text information.

[0048] Specifically, if it is recognized that the audio information contains voice data, text information corresponding to the voice data in the audio information and the corresponding recognition time are obtained.

[0049] See also Figure 3 , Figure 3 This is the storage structure of the data queue FIFO provided by the present invention. In a specific embodiment, the text information of the voice data includes at least one word text. Each word text W contained in the text information of the voice data is recorded, and the recognition time T of each word text in the text information of the voice data is recorded. The recognized word text W and the recognition time T of the recognized word text are saved in the data queue FIFO in sequence; in response to the queue length of the data queue FIFO reaching a preset length, the recognition time of the last word text saved in the data queue is determined as the recognition time of the text information.

[0050] In a specific embodiment, n recognized word texts are input into a data queue FIFO. The data queue FIFO adopts a first-in-first-out data structure. FIFO1 to FIFOn correspond to audio channels 1 to n respectively and are used to store the speech recognition texts of the corresponding audio channels. For example, the length of each data queue is uniformly 5, which means that the data queue can store the most recent 5 speech recognition results. The speech recognition results include the recognized word text W. i And the recognition time Ti of the word text. Among them, Index is the position number of the element of each queue. The head of the queue is Index = 1, which stores the latest result. And so on, to the tail of the queue is Index = 5. The queue uses a first-in-first-out method to store speech recognition results (Wi, Ti). If a new text is added to the queue, the content of Index = 1 to 4 will be moved to Index 2 to 5 respectively, and the original data of Index = 5 will be discarded. Then the new result is placed at Index = 1.

[0051] The recognition time is the time when the voice data is recognized. In other words, the recognition time corresponding to each audio channel is the starting time when the same voice data collected by the corresponding audio channel is recognized.

[0052] Through this step, the text information in the audio information corresponding to each audio channel and the corresponding recognition time can be identified.

[0053] S206: Determine whether the text information stored in the data queues of preset lengths corresponding to the respective audio channels is consistent.

[0054] Specifically, in the present embodiment, each audio channel in the data queue FIFO stores 5 speech recognition results, and the 5 word texts W in each data queue FIFO are strung together into a row of text in sequence. Since the mixed audio source is all in the same classroom, when someone speaks in the classroom at a volume exceeding the preset decibel, each audio channel will collect the voice, and the corresponding voice recognition module will recognize it and output the recognition result. Although there may be a delay in each audio channel, there will definitely be pauses and gaps in normal speech. Therefore, there will always be a situation where the same text information is stored in each audio channel in all data queue FIFOs at a certain moment. For example, when a teacher is giving a lecture, there will definitely be a pause of more than 5 seconds after finishing a sentence. Although there are delay differences between the audio channels, the general time difference will not exceed 2 seconds. After the pause of more than 5 seconds, the audio information of all audio channels should have collected the teacher's speech 5 seconds ago and sent it to the voice recognition module for recognition.

[0055] In order to determine the recognition time corresponding to each audio channel based on the same voice data, it is necessary to determine whether the recognized text information corresponding to each audio channel is consistent.

[0056] If the text information recognized by each audio channel is consistent, the process directly jumps to step S208; if the text information recognized by each audio channel is inconsistent, the process directly jumps to step S202.

[0057] S207: Determine the number of audio frames buffered in the queue corresponding to each audio channel according to the recognition time of the same word text in the text information corresponding to each audio channel.

[0058] Specifically, when the same text information is stored in each data queue FIFO, it means that the text information corresponding to each audio channel is the same. Then, the recognition time corresponding to each audio channel is calculated based on the recognition time corresponding to the same word text in the text information.

[0059] In a specific embodiment, the recognition time of the first word text in the text information corresponding to each audio channel is determined, that is, the recognition time T of the results of all queues with Index=5 is taken, and is recorded as T1'-Tn'.

[0060] S208: Selecting the latest recognition time among the recognition times corresponding to at least two audio channels.

[0061] Specifically, the recognition times corresponding to different audio channels are obtained through the above steps, and the recognition times corresponding to the different audio channels are sorted from largest to smallest, and the largest recognition time is selected as the latest recognition time corresponding to the multiple audio channels. For example, the recognition times of the audio channels are sorted as follows: DavgT0 (channel 0), DavgT1 (channel 1), ..., DavgTn (channel n), and the largest recognition time is selected: Tm'=Max{DavgT1', ..., DavgTn'}, where m is the channel number with the largest corresponding time T.

[0062] S209: Calculate the difference between the recognition time corresponding to each audio channel and the latest recognition time to determine the number of audio frames buffered in the queue corresponding to each audio channel.

[0063] Specifically, the difference between the recognition time corresponding to the audio channel and the latest recognition time is calculated; based on the difference, the sampling rate of the audio information and the number of audio samples in each audio frame, the number of queue cache audio frames corresponding to each audio channel is determined.

[0064] In one embodiment, the recognition times of other audio channels are adjusted based on the latest recognition time. That is, the recognition times of other audio channels need to be consistent with the latest recognition time of audio channel m. The difference between the recognition time corresponding to each audio channel and the latest recognition time is calculated.

[0065] According to the difference between the recognition time corresponding to the audio channel and the latest recognition time and the sampling rate of the audio information, the number of audio frames buffered in the queue corresponding to the audio channel is determined. i ', that is, L′ i =(Tm′-Ti′)*f i / N i Among them, L i is a positive integer, i is the audio channel number, f i Get the sampling rate of audio information for the audio channel, N i The number of audio samples in each audio frame of the audio channel. Specifically, the number of audio frames cached in the queue represents the number of audio frames pre-stored in the queue, and each audio frame contains the same number of audio samples. In this embodiment, the audio sample can be blank data, that is, empty sound data.

[0066] S210: Storing a preset number of audio frames in a corresponding queue of each audio channel.

[0067] Specifically, to reduce the latency differences between audio channels and save performance, a preset number of audio frames can be stored in each audio channel. The preset number of audio frames stored in each audio channel can be the same or different. In one specific embodiment, the preset number of audio frames stored in each audio channel is the same. Furthermore, the number of audio frames stored in each audio channel is L, and the duration of each of the L audio frames is equal.

[0068] S211: Determine whether the preset number of audio frames stored in the queue of the audio channel is greater than the number of audio frames buffered in the queue of the corresponding audio channel.

[0069] Specifically, the preset number of audio frames stored in the queue of the audio channel is compared with the number of audio frames buffered in the queue of the corresponding audio channel.

[0070] If the preset number of audio frames stored in the queue of the audio channel is less than the number of audio frames cached in the queue of the corresponding audio channel, jump directly to step S212; if the preset number of audio frames stored in the queue of the audio channel is greater than the number of audio frames cached in the queue of the corresponding audio channel, jump directly to step S213.

[0071] S212: adding preset audio frames so that the number of audio frames stored in the queue of the audio channel reaches the number of audio frames buffered in the queue of the corresponding audio channel.

[0072] Specifically, in response to the preset number of audio frames stored in the queue of the audio channel being less than the number of audio frames cached in the queue of the corresponding audio channel, the preset audio frames are added to make the number of audio frames stored in the queue reach the number of audio frames cached in the queue of the corresponding audio channel.

[0073] S213: discarding preset audio frames so that the number of audio frames stored in the queue of the audio channel reaches the number of audio frames buffered in the queue of the corresponding audio channel.

[0074] Specifically, in response to the preset number of audio frames stored in the queue of the audio channel being greater than the number of audio frames cached in the queue of the corresponding audio channel, the preset audio frames are discarded so that the number of audio frames stored in the queue reaches the number of audio frames cached in the queue corresponding to the audio channel.

[0075] In another embodiment, the calculated preset number of audio frames buffered by the queue corresponding to each audio channel may be directly added to the queue of each audio channel.

[0076] The present embodiment provides a delayed transmission method for audio data, comprising obtaining the same audio information transmitted by at least two audio channels respectively; the audio information includes voice data; performing voice recognition on the audio information transmitted by the at least two audio channels respectively, and determining the recognition time of the voice data; and determining the number of queue-cached audio frames corresponding to the at least two audio channels respectively based on the recognition time corresponding to the at least two audio channels. The present application identifies the audio information transmitted by different audio channels, and determines the number of queue-cached audio frames of the audio channels based on the recognition time corresponding to all audio channels respectively, so that the mixing module can receive audio data transmitted synchronously by different audio channels, thereby improving the mixing effect.

[0077] See also Figure 4 , Figure 4 : is a schematic block diagram of the delayed transmission device for audio data provided by the present invention. This embodiment provides a delayed transmission device 40 for audio data, which includes an acquisition module 41, an identification module 42, and a processing module 43. The acquisition module 41 is used to acquire the same audio information transmitted by at least two audio channels; the audio information includes voice data; the identification module 42 is used to perform voice recognition on the audio information transmitted by the at least two audio channels and determine the recognition time of the voice data; the processing module 43 is used to determine the number of audio frames in the queue buffer corresponding to the at least two audio channels based on the recognition time corresponding to the at least two audio channels.

[0078] In one embodiment, the recognition module 42 is in a running mode during the calibration phase for the number of audio frames buffered in the queue corresponding to the audio channel. After the calibration is complete, the recognition module 42 is in a bypass mode. When in the bypass mode, the recognition module 42 does not process the input audio data and directly passes it to the next-level module.

[0079] The delayed transmission device for audio data provided in this embodiment identifies the audio information transmitted by different audio channels and determines the number of audio frames to be cached in the queue of the audio channel based on the identification time corresponding to all audio channels, so that the mixing module can receive audio data transmitted synchronously by different audio channels, thereby improving the mixing effect.

[0080] See Figure 5 , Figure 5 This is a schematic block diagram of an embodiment of a terminal provided by the present invention. Terminal 70 in this embodiment includes a processor 71, a memory 72, and a computer program stored in memory 72 and executable by processor 71. When executed by processor 71, this computer program implements the aforementioned method for delayed transmission of audio data. To avoid repetition, detailed descriptions are omitted here.

[0081] See Figure 6 , Figure 6 It is a schematic block diagram of an embodiment of a computer-readable storage medium provided by the present invention.

[0082] In an embodiment of the present application, a computer-readable storage medium 90 is further provided. The computer-readable storage medium 90 stores a computer program 901. The computer program 901 includes program instructions. The processor executes the program instructions to implement the delayed transmission method of audio data provided in an embodiment of the present application.

[0083] The computer-readable storage medium 90 may be an internal storage unit of the computer device described in the aforementioned embodiment, such as a hard disk or memory of the computer device. The computer-readable storage medium 90 may also be an external storage device of the computer device, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash memory card, etc.

[0084] The above are merely embodiments of the present invention and are not intended to limit the scope of patent protection of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention's description and drawings, or directly or indirectly applied in other related technical fields, are also included in the scope of patent protection of the present invention.

Claims

1. A delayed transmission method for audio data, characterized in that: The delayed transmission method of audio data comprises: Acquire the same audio information transmitted respectively by at least two audio channels; the audio information includes voice data; performing voice recognition on the audio information transmitted through the at least two audio channels respectively, and determining a recognition time of the voice data; Determining, based on the recognition times respectively corresponding to the at least two audio channels, the number of queue buffered audio frames respectively corresponding to the at least two audio channels; The determining, based on the recognition times respectively corresponding to the at least two audio channels, the number of queue buffered audio frames respectively corresponding to the at least two audio channels includes: Selecting the latest recognition time among the recognition times corresponding to the at least two audio channels; The difference between the recognition time corresponding to each of the audio channels and the latest recognition time is calculated respectively to determine the number of audio frames buffered in the queue corresponding to each of the audio channels.

2. The delayed transmission method of audio data according to claim 1, characterized in that: The performing speech recognition on the audio information transmitted through the at least two audio channels respectively and determining the recognition time of the speech data includes: performing speech recognition on the audio information; Determining whether the audio information recognizes text information of the voice data; In response to recognizing the text information of the voice data, a recognition time of the recognized text information is determined; wherein the recognition time is the moment when the text information corresponding to the voice data is recognized.

3. The delayed transmission method of audio data according to claim 2, characterized in that: The text information includes at least one word text; In response to recognizing the text information of the voice data, determining a recognition time of the recognized text information includes: The recognized word text and the recognition time of the recognized word text are sequentially saved in a data queue; In response to the queue length of the data queue reaching a preset length, the recognition time of the word text last saved in the data queue is determined as the recognition time of the text information.

4. The method for delayed transmission of audio data according to claim 3, wherein: The determining, based on the recognition times respectively corresponding to the at least two audio channels, the number of queue buffered audio frames respectively corresponding to the at least two audio channels includes: Determining whether the text information stored in the data queues of the preset length corresponding to the respective audio channels is the same; In response to the fact that the text information stored in the data queues of the preset length corresponding to the respective audio channels is the same, the number of queue cache audio frames corresponding to the respective audio channels is determined based on the recognition time of the same word text in the text information corresponding to the respective audio channels.

5. The delayed transmission method of audio data according to claim 1, characterized in that: The calculating the difference between the recognition time corresponding to each of the audio channels and the latest recognition time to determine the number of queue buffered audio frames corresponding to each of the audio channels includes: Calculating the difference between the recognition time corresponding to the audio channel and the latest recognition time; The number of audio frames buffered in the queue of the audio channel is determined according to the difference, a sampling rate of the audio information, and the number of audio samples in each audio frame in the audio information.

6. The delayed transmission method of audio data according to claim 1, characterized in that: The method further comprises: determining the number of queue buffered audio frames corresponding to the at least two audio channels based on the recognition times corresponding to the at least two audio channels respectively; and then: Adding a preset number of audio frames corresponding to the queue buffer audio frames to the queue of each audio channel.

7. A delayed transmission device for audio data, characterized in that: The delayed transmission device of the audio data comprises: An acquisition module, configured to acquire the same audio information transmitted respectively by at least two audio channels; the audio information includes voice data; a recognition module, configured to perform voice recognition on the audio information transmitted through the at least two audio channels respectively, and determine a recognition time of the voice data; The processing module is configured to determine, based on the identification times corresponding to the at least two audio channels, the number of queue-cached audio frames corresponding to each of the at least two audio channels; select a latest identification time among the identification times corresponding to the at least two audio channels; and calculate the difference between the identification time corresponding to each of the audio channels and the latest identification time to determine the number of queue-cached audio frames corresponding to each of the audio channels.

8. A terminal, characterized in that: The terminal includes a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor is configured to execute program data to implement the steps of the method for delayed transmission of audio data according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the delayed transmission method of audio data according to any one of claims 1 to 6 are implemented.

Images