A simple implementation method for APP voice broadcast

Abstract

The present application discloses a simple implementation method for APP voice broadcast, which includes determining multiple target audio segments to be processed and corresponding voice playback timing information based on an acquired voice broadcast setting instruction; decoding multiple target audio segments to obtain corresponding multiple target audio signals; performing audio preprocessing on multiple target audio signals, and setting the audio feature parameters of all target audio signals within a unified numerical range; determining the splicing points between multiple target audio signals, splicing and combining the multiple target audio signals according to the splicing points to obtain an initial voice broadcast signal; the splicing points in the initial voice broadcast signal meet the connection between the end sampling point of the previous target audio signal in the multiple target audio signals and the starting sampling point of the next target audio signal; performing broadcast voice quality judgment on the initial voice broadcast signal; when the initial voice broadcast signal meets the voice optimization processing conditions in the preset voice optimization judgment rule, optimizing the initial voice broadcast signal to obtain a target voice broadcast signal; encoding the target voice broadcast signal to obtain a target audio broadcast file; the present application has the effect of providing a simple implementation method for text-to-speech broadcast with excellent sound quality and conducive to saving enterprise operating costs when facing the demand for APP voice broadcast in dialects other than Mandarin.

Description

Technical Field

[0001] The present application relates to the field of voice broadcast technology, and in particular to a simple method for implementing APP voice broadcast. Background Art

[0002] Currently, in the APP development industry, the implementation of text-to-speech (TTS) function mainly relies on the system TTS engine built into the mobile device and the third-party TTS SDK (Software Development Kit) service; system-level TTS is generally pre-installed by the operating system provider, which can meet basic speech synthesis needs, but has limited support for non-Mandarin dialects (such as Cantonese), and its speech synthesis effect is not satisfactory. There is also room for improvement in the voice style and naturalness of non-Mandarin dialects.

[0003] SDKs such as those provided by TTS service providers usually have a richer range of languages, dialect support and better sound quality, which can bring a more personalized and realistic speech synthesis experience; however, such solutions are for commercial use, and developers often need to pay corresponding fees based on the number of API (Application Programming Interface) calls. For those apps that frequently call TTS services, the accumulated cost expenditure will increase significantly, which means that as the number of app users increases and the frequency of use increases, the related expenses of the voice broadcast function will increase proportionally, which will undoubtedly increase the operating burden of the enterprise.

[0004] Based on this, in response to the demand for APP voice broadcast in non-Mandarin dialects, a simple implementation method for text-to-speech broadcasting with excellent sound quality and conducive to saving enterprise operating costs has practical application value and significance. Summary of the invention

[0005] In order to provide a simple implementation method for text-to-speech broadcasting with excellent sound quality and conducive to saving enterprise operating costs when facing the demand for APP voice broadcasting in dialects other than Mandarin, the present application provides a simple implementation method for APP voice broadcasting.

[0006] In the first aspect, the invention objective of the present application is achieved by adopting the following technical solutions:

[0007] A simple method for implementing APP voice broadcasting, including:

[0008] Determine multiple target audio segments to be processed and corresponding voice playback timing information based on the acquired voice broadcast setting instruction; decode the multiple target audio segments to obtain corresponding multiple target audio signals;

[0009] Performing audio preprocessing on the plurality of target audio signals, and setting the audio feature parameters of all the target audio signals within a uniform value range;

[0010] Determine the splicing points between the multiple target audio signals according to the voice playback timing information, splice the multiple target audio signals according to the splicing points to obtain an initial voice broadcast signal; the splicing point in the initial voice broadcast signal satisfies the connection between the end sampling point of the previous target audio signal and the start sampling point of the next target audio signal among the multiple target audio signals;

[0011] The initial voice broadcast signal is subjected to a voice quality judgment according to a preset voice optimization judgment rule; when the initial voice broadcast signal satisfies the voice optimization processing condition in the preset voice optimization judgment rule, the initial voice broadcast signal is optimized to obtain a target voice broadcast signal;

[0012] The target voice broadcast signal is encoded to obtain a target audio broadcast file.

[0013] By adopting the above technical solution, the present application can provide personalized customization and a realistic speech synthesis experience. Users can merge and generate audio files by inputting voice broadcast setting instructions. The present application can flexibly adapt to various usage scenarios, such as APP cashier broadcast, takeaway order broadcast, restaurant queue call broadcast, etc., and realize efficient speech synthesis and real-time broadcast services by integrating sound materials containing rich and diverse voice styles, intonations and speaking speeds.

[0014] Specifically, the multiple target audio clips to be processed are audio clips recorded in non-Mandarin dialects, which can be prepared and stored in advance according to actual scenarios. When facing the demand for APP voice broadcast in non-Mandarin dialects, in order to facilitate the optimization processing of the target audio clips and the quality judgment of the audio signal, each target audio clip is first decoded, that is, decoded into a digitized target audio signal, and then each target audio signal is audio preprocessed. The audio preprocessing includes making the multiple target audio signals have the same sampling rate, quantization bit depth and number of channels and / or volume matching processing for target audio signals with inconsistent volumes, so as to effectively ensure that the multiple target audio signals after splicing and merging can be played normally on the playback device with uniform sound quality, and the volume of the splicing point can be smoothly transitioned; the voice playback timing information includes the playback timing of the multiple target audio signals and the positions of each splicing point. The voice playback timing information can be used to determine the appropriate splicing position, such as splicing in silent passages, repeated rhythm points or melody transitions, so as to effectively reduce the sense of discontinuity in listening.

[0015] In the next step, according to the voice playback timing information and the determined multiple splicing points, multiple target audio signals are spliced ​​to obtain an initial voice broadcast signal. After the splicing and merging are completed, in order to further improve the sound quality of the text-to-speech broadcast and determine whether there are obvious splicing traces, sound quality degradation and other problems, the present application further uses the voice optimization judgment rule to evaluate the voice quality. When the initial voice broadcast signal meets the voice optimization processing conditions, the initial voice broadcast signal is optimized, such as fine-tuning the splicing points according to the actual broadcast needs to obtain the target voice broadcast signal, and finally the target voice broadcast signal is re-encoded to obtain the target audio broadcast file in the audio format specified by the user; the present application uses an efficient voice and voice quality monitoring method to not only help solve the personalized broadcast needs for local dialects or special languages ​​in mobile applications (APPs), but also significantly reduce the high cost pressure caused by the frequent use of high-quality voice services, that is, to provide a simple implementation method for text-to-speech broadcast with excellent sound quality and conducive to saving enterprise operating costs.

[0016] In a preferred example of the present application, a database pre-stored with multiple types of audio clip samples includes a language type, an intonation type, a speech speed type, and a timbre type, and each audio clip sample is associated with an audio feature identifier; the method of determining multiple target audio clips to be processed and corresponding voice playback timing information based on the acquired voice broadcast setting instruction includes:

[0017] Based on the acquired voice broadcast setting instruction, the corresponding text information to be broadcast and the audio characteristic parameters are identified, and based on the playback timing of the text information to be broadcast, the corresponding voice playback timing information is obtained;

[0018] Based on the text information to be broadcast and the audio feature parameters, multiple audio segment samples are extracted from the database according to the audio feature identifier of each audio segment sample; and the extracted multiple audio segment samples are determined as target audio segments to be processed.

[0019] By adopting the above technical solution, when facing the demand for APP voice broadcast in dialects other than Mandarin, in order to facilitate the simple implementation of APP voice broadcast, a variety of audio segment samples are first pre-collected and stored in the database of the audio operating system (such as the mobile phone system API). The various types of audio segment samples include audio segment samples of different languages, different intonations, different speech speeds and different timbres (as audio segment materials to be spliced ​​and merged), and the audio feature identifier is a unique identification code of the broadcast text information and audio feature parameter information of the audio segment sample; specifically, based on the voice broadcast setting set by the user, The playback timing of the text information to be broadcast of the instruction obtains the corresponding voice playback timing information. For example, the text information to be broadcast is the Cantonese voice and playback order of "Payment successful, amount 10.5 yuan", then in actual application, multiple audio clip samples of "collection", "success", "amount", "10", "point", "5" and "yuan" in Cantonese can be extracted from the database as the target audio clips to be processed, so as to facilitate subsequent signal preprocessing and voice signal splicing and merging, thereby providing a simple and easy-to-implement APP voice broadcast method, and also providing users with voice interaction services that are more in line with actual needs.

[0020] In a preferred example of the present application, the audio feature parameters include audio sampling rate, quantization bit depth and number of audio channels; the extracting multiple audio segment samples from the database based on the to-be-broadcast text information and the audio feature parameters according to the audio feature identifier of each audio segment sample specifically includes:

[0021] In the database, the audio segment samples whose text information to be broadcast matches the text information to be broadcast are screened to obtain a plurality of audio segment samples to be confirmed;

[0022] According to the audio feature identifier of each audio segment sample, identifying and obtaining the set audio feature parameter values ​​of the plurality of audio segment samples to be confirmed;

[0023] Determine an extraction parameter range of each audio feature parameter of each audio segment sample to be extracted based on the audio sampling rate, quantization bit depth and number of audio channels set in the voice broadcast setting instruction;

[0024] Based on the extraction parameter range of each audio feature parameter of each audio segment sample, audio segment samples whose set audio feature parameter values ​​all conform to the corresponding extraction parameter range are extracted respectively.

[0025] By adopting the above technical scheme, when extracting multiple audio clip samples from the database, firstly preliminarily screen the multiple audio clip samples whose text information to be broadcast is consistent with the text information to be broadcast (i.e., the multiple audio clip samples to be confirmed), that is, preliminarily filter out the audio clip samples whose content information to be broadcast (i.e., the text information) is inconsistent, and then identify and obtain the set audio feature parameter values ​​of the multiple audio clip samples to be confirmed. Since the audio feature parameter values ​​of the audio clip sample materials are set within the specified numerical range when the materials of each audio clip sample are recorded in the early stage, the extraction parameter range requirements of each audio feature parameter to be extracted (referring to the audio sampling rate, quantization bit depth and number of audio channels) are confirmed, and then the audio clip samples are screened according to the set audio feature parameter values ​​of each audio feature parameter of each audio clip sample to obtain multiple audio clip samples whose audio feature parameters all meet the requirements of the voice broadcast setting instruction. At this time, the audio sampling rate, quantization bit depth and number of audio channels of the multiple audio clip samples screened out are consistent, which is conducive to ensuring that the spliced ​​audio can be played normally on the playback device with uniform sound quality, and the screening and extraction of audio clip samples is simple, fast and efficient.

[0026] In a preferred example of the present application, the audio preprocessing of the plurality of target audio signals to set the audio feature parameters of all the target audio signals within a uniform value range includes:

[0027] Performing volume intensity detection on the plurality of target audio signals to obtain corresponding volume intensities;

[0028] When the volume intensity of a target audio signal exceeds a preset volume intensity threshold, normalizing the target audio signal;

[0029] or,

[0030] Traversing and collecting the audio tracks of the target audio segments during the broadcasting process to obtain corresponding sampling values;

[0031] The gain factor of each target audio signal is calculated according to the preset volume intensity threshold and the corresponding sampling value; the audio track played by the target audio segment corresponding to all the target audio signals is multiplied by the gain factor to obtain the corresponding adjusted audio track value, so as to perform gain processing on the target audio signal.

[0032] By adopting the above technical solution, for target audio clips with inconsistent volumes, the present application provides a matching strategy for consistent volume matching by means of normalization processing or gain adjustment. Multiple target audio clips subjected to normalization processing and / or gain adjustment are beneficial for smooth transition of audio clips after splicing and merging, thereby facilitating improving the quality of audio splicing.

[0033] In a preferred example, the present application also includes:

[0034] According to a preset silent segment filtering rule, multiple target audio signals after normalization or gain processing are traversed and collected to obtain multiple sampling point values;

[0035] When the sampling point value of a target audio signal meets the preset silence segment filtering rule, the corresponding silence segment data stream of the target audio signal is filtered and removed; the preset silence segment filtering rule satisfies that the volume intensity values ​​of the sampling points of multiple consecutive frames are all lower than the preset silence volume threshold.

[0036] By adopting the above technical solution, a traversal method is used to detect whether there are silent segments in multiple target audio signals. That is, when the sampling point value of a certain target audio signal meets the filtering condition of the silent segment filtering rule, the sampling point value at this time is the volume intensity value of the sampling point, which represents that the signal data stream where the sampling point is located meets the volume intensity values ​​of the sampling points detected for multiple consecutive frames are all lower than the set silent volume threshold, thereby realizing the automatic removal of silent segments in each target audio signal. Only identification and filtering technology is used, thereby improving the compactness and continuity of APP voice broadcasting.

[0037] In a preferred example, the present application also includes:

[0038] When the plurality of target audio signals are spliced ​​together according to the splicing point to obtain an initial voice broadcast signal, a volume intensity of the splicing point in the initial voice broadcast signal is faded in according to a preset fade-in algorithm, and a volume intensity of the splicing point in the initial voice broadcast signal is faded out according to a preset fade-out algorithm;

[0039] The preset fade-in algorithm is shown in formula (1):

[0040] G(x)=Q(x)*e cx+b +δ*Y x (1)

[0041] Where G(x) is the volume intensity at time x, x is the time change of the preset time period for fade-in processing, Q(x) is the volume intensity of the splicing point, c and b are constants, δ is the adjustment coefficient, and Y x The calibrated volume intensity at time point x.

[0042] By adopting the above technical solution, the target audio signal is a digitized audio data stream. By adopting the smooth transition splicing processing of fade-in and fade-out processing, the transition between the target audio signals at both ends is natural and not abrupt. That is, the six target audio signal streams of "successful collection", "amount", "10", "points", "5" and "yuan" are smoothly spliced ​​together in sequence, ensuring that the logic of the broadcast content is clear and the transition is natural, giving users a listening experience like a real-time broadcast by a real person.

[0043] In a preferred example of the present application, the preset fade-out algorithm is as shown in formula (2):

[0044] F(x)=Q(x)*e -(kx+d) +δ*Y x (2)

[0045] Where F(x) is the volume intensity at time x, x is the time change of the preset time period for fade-out processing, Q(x) is the volume intensity at the splicing point, k and d are constants, δ is the adjustment coefficient, and Y x The calibrated volume intensity at time point x.

[0046] By adopting the above technical solution, formula (2) is helpful to achieve a fade-out processing effect at a set splicing point.

[0047] In the second aspect, the invention objective of the present application is achieved by adopting the following technical solutions:

[0048] A simple implementation system for APP voice broadcast, the system includes:

[0049] An audio segment acquisition and decoding module is used to determine multiple target audio segments to be processed and corresponding voice playback timing information based on the acquired voice broadcast setting instruction; decode multiple target audio segments to obtain corresponding multiple target audio signals; an audio signal preprocessing module is used to perform audio preprocessing on the multiple target audio signals, and set the audio feature parameters of all the target audio signals within a unified value range;

[0050] An audio segment splicing and merging module is used to determine the splicing points between the multiple target audio signals according to the voice playback timing information, and splice and merge the multiple target audio signals according to the splicing points to obtain an initial voice broadcast signal; the splicing point in the initial voice broadcast signal satisfies the connection between the end sampling point of the previous target audio signal and the starting sampling point of the next target audio signal among the multiple target audio signals;

[0051] The audio quality judgment and optimization module is used to judge the broadcast voice quality of the initial voice broadcast signal according to the preset voice optimization judgment rule; when the initial voice broadcast signal meets the voice optimization processing conditions in the preset voice optimization judgment rule, the initial voice broadcast signal is optimized to obtain the target voice broadcast signal;

[0052] The audio file encoding module is used to encode the target voice broadcast signal to obtain a target audio broadcast file.

[0053] By adopting the above technical solution, the multiple target audio clips to be processed are audio clips recorded in non-Mandarin dialects, which can be prepared and stored in advance according to actual scenarios. When facing the APP voice broadcast requirements of non-Mandarin dialects, in order to facilitate the optimization processing of the target audio clips and the quality judgment of the audio signal, each target audio clip is first decoded, that is, decoded into a digitized target audio signal, and then each target audio signal is audio preprocessed. The audio preprocessing includes making the multiple target audio signals have the same sampling rate, quantization bit depth and number of channels and / or volume matching processing for target audio signals with inconsistent volumes, so as to effectively ensure that the multiple target audio signals after splicing and merging can be played normally on the playback device with uniform sound quality, and the volume of the splicing point is smoothly transitioned; the voice playback timing information includes the playback timing of the multiple target audio signals and the positions of each splicing point. The voice playback timing information can be used to determine the appropriate splicing position, such as splicing in silent passages, repeated rhythm points or melody transitions, so as to effectively reduce the sense of discontinuity in listening.

[0054] In the next step, according to the voice playback timing information and the determined multiple splicing points, multiple target audio signals are spliced ​​to obtain an initial voice broadcast signal. After the splicing and merging are completed, in order to further improve the sound quality of the text-to-speech broadcast and determine whether there are obvious splicing traces, sound quality degradation and other problems, the present application further uses the voice optimization judgment rule to evaluate the voice quality. When the initial voice broadcast signal meets the voice optimization processing conditions, the initial voice broadcast signal is optimized, such as fine-tuning the splicing points according to the actual broadcast needs to obtain the target voice broadcast signal, and finally the target voice broadcast signal is re-encoded to obtain the target audio broadcast file in the audio format specified by the user; the present application uses an efficient voice and voice quality monitoring method to not only help solve the personalized broadcast needs for local dialects or special languages ​​in mobile applications (APPs), but also significantly reduce the high cost pressure caused by the frequent use of high-quality voice services, that is, to provide a simple implementation system with excellent sound quality for text-to-speech broadcasts and conducive to saving enterprise operating costs.

[0055] In the third aspect, the invention objective of this application is achieved by adopting the following technical solutions:

[0056] A smart device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the simple method for implementing the above-mentioned APP voice broadcast are implemented.

[0057] In a fourth aspect, the invention objective of the present application is achieved by adopting the following technical solutions:

[0058] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned simple implementation method of APP voice broadcast.

[0059] In summary, the present application includes at least one of the following beneficial technical effects:

[0060] 1. The multiple target audio clips to be processed are audio clips recorded in non-Mandarin dialects, which can be prepared and stored in advance according to actual scenarios. When facing the demand for APP voice broadcast in non-Mandarin dialects, in order to facilitate the optimization of the target audio clips and the quality judgment of the audio signal, each target audio clip is first decoded, that is, decoded into a digitized target audio signal, and then audio preprocessing is performed on each target audio signal. The audio preprocessing includes making the multiple target audio signals have the same sampling rate, quantization bit depth and number of channels and / or performing volume matching on target audio signals with inconsistent volumes, so as to effectively ensure that the multiple target audio signals after splicing and merging can be played normally on the playback device with uniform sound quality, and the volume of the splicing point can be smoothly transitioned; the voice playback timing information includes the playback timing of the multiple target audio signals and the positions of each splicing point. The voice playback timing information can be used to determine the appropriate splicing position, such as splicing at a silent section, a repeated rhythm point or a melody transition, so as to effectively reduce the sense of discontinuity in the listening experience;

[0061] Next, according to the voice playback timing information and the determined multiple splicing points, multiple target audio signals are spliced ​​to obtain an initial voice broadcast signal. After the splicing and merging are completed, in order to further improve the sound quality of the text-to-speech broadcast and determine whether there are obvious splicing traces, sound quality degradation and other problems, the present application further uses the voice optimization judgment rule to evaluate the voice quality. When the initial voice broadcast signal meets the voice optimization processing conditions, the initial voice broadcast signal is optimized, such as fine-tuning the splicing points according to the actual broadcast demand to obtain the target voice broadcast signal, and finally the target voice broadcast signal is re-encoded to obtain the target audio broadcast file in the audio format specified by the user; the present application not only helps to solve the personalized broadcast demand for local dialects or special languages ​​in mobile applications (APPs) through an efficient voice and voice quality monitoring method, but also can significantly reduce the high cost pressure caused by the frequent use of high-quality voice services, that is, to provide a simple implementation system with excellent sound quality for text-to-speech broadcast and conducive to saving enterprise operating costs;

[0062] 2. Pre-collect audio clip samples of various types and store them in a database of an audio operating system (such as a mobile phone system API). The various types of audio clip samples include audio clip samples of different languages, different intonations, different speech speeds and different timbres (as audio clip materials to be spliced ​​and merged). The audio feature identifier is a unique identification code of the broadcast text information and audio feature parameter information of the audio clip sample; specifically, the corresponding voice playback timing information is obtained based on the playback timing of the text information to be broadcast of the voice broadcast setting instruction set by the user. For example, if the text information to be broadcast is the Cantonese voice and playback order of "successful payment, amount 10.5 yuan", in actual application, multiple audio clip samples of "collection", "success", "amount", "10", "points", "5" and "yuan" in Cantonese can be extracted from the database as target audio clips to be processed, so as to facilitate subsequent signal preprocessing and voice signal splicing and merging, thereby providing a simple and easy-to-implement APP voice broadcast method and providing users with voice interaction services that are more in line with actual needs;

[0063] 3. For target audio clips with inconsistent volumes, the present application provides a matching strategy for consistent volume matching by means of normalization processing or gain adjustment. Multiple target audio clips subjected to normalization processing and / or gain adjustment are beneficial for smooth transition of audio clips after splicing and merging, thereby facilitating improved audio splicing quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 This is a flow chart of a simple method for implementing APP voice broadcast in one embodiment of the present application;

[0065] Figure 2 This is a flowchart of step S1 in a simple method for implementing APP voice broadcast in one embodiment of the present application;

[0066] Figure 3 This is a flowchart of step S12 in a simple method for implementing APP voice broadcast in one embodiment of the present application. DETAILED DESCRIPTION

[0067] The present application is further described in detail below in conjunction with the accompanying drawings.

[0068] In one embodiment, if Figure 1 As shown, the present application discloses a simple method for implementing APP voice broadcast, which specifically includes the following steps:

[0069] S1: determining a plurality of target audio segments to be processed and corresponding voice play timing information based on the acquired voice broadcast setting instruction; decoding the plurality of target audio segments to obtain a plurality of corresponding target audio signals.

[0070] In this embodiment, the voice broadcast setting instruction is a designation of the voice broadcast setting input by the user through a terminal, and the terminal includes an API application (APP audio processing application) in a mobile phone, a computer, and a web page.

[0071] Specifically, the decoded target audio signal is an audio data stream in a PCM (Pulse Code Modulation) format.

[0072] For example, for the application scenario of APP voice broadcast of store cashier broadcast, when constructing a Cantonese voice notification to broadcast successful payment, this embodiment takes the file format of the target audio segment as .wav as an example, and first needs to filter out the following six pre-recorded Cantonese audio materials:

[0073] 1) A complete voice .wav file of "Payment received successfully" to clearly inform the transaction status;

[0074] 2) A separate .wav file of "amount" as a guide to the amount value;

[0075] 3) The digital pronunciation .wav file of "10" is used to announce the ten digits of a specific amount;

[0076] 4) "dot" .wav file, used to connect the integer and decimal parts;

[0077] 5) The digital pronunciation of "5".wav file, which broadcasts the decimal places of the amount;

[0078] 6) The .wav file of "yuan" is used as a currency unit identifier.

[0079] S2: performing audio preprocessing on the multiple target audio signals, and setting the audio feature parameters of all the target audio signals within a uniform value range.

[0080] In this embodiment, the audio feature parameters include sampling rate, quantization bit depth and number of channels; the audio preprocessing includes screening the audio feature parameters of multiple target audio signals so that the screened multiple target audio signals have the same sampling rate, quantization bit depth and number of channels, and then matching the target audio signals with inconsistent volumes, such as performing audio processing adjustments such as normalization and gain adjustment.

[0081] S3: Determine the splicing points between multiple target audio signals according to the voice playback timing information, splice the multiple target audio signals according to the splicing points and obtain an initial voice broadcast signal; the splicing points in the initial voice broadcast signal meet the requirements that the end sampling point of the previous target audio signal among the multiple target audio signals is connected with the starting sampling point of the next target audio signal.

[0082] In this embodiment, the voice playback timing information includes the playback timing of multiple target audio signals and the positions of each splicing point; the voice playback timing information provides a splicing order for the multiple target audio signals to determine the arrangement order of the splicing points between the multiple target audio signals.

[0083] Specifically, the appropriate splicing position can be determined through the voice playback timing information, such as splicing at silent passages, repeated rhythm points or melody transitions, so as to effectively reduce the sense of discontinuity in listening.

[0084] S4: judging the voice quality of the initial voice broadcast signal according to the preset voice optimization judgment rule; when the initial voice broadcast signal meets the voice optimization processing conditions in the preset voice optimization judgment rule, optimizing the initial voice broadcast signal to obtain the target voice broadcast signal.

[0085] In this embodiment, the voice optimization judgment rule includes judging the voice silence segment. By playing and monitoring, it is judged whether the initial voice broadcast signal after splicing has obvious splicing traces, sound quality degradation and other problems, such as judging the difference in volume intensity of each splicing point.

[0086] Specifically, when the initial voice broadcast signal meets the voice optimization processing conditions in the preset voice optimization judgment rule, that is, the initial voice broadcast signal has a silent segment, the corresponding silent segment of the initial voice signal in PCM format is deleted and filtered.

[0087] S5: Encode the target voice broadcast signal to obtain a target audio broadcast file.

[0088] In this embodiment, the target voice broadcast signal is encoded and compressed according to a specified file format (such as MP3, WAV, AAC and other audio formats) and then output to obtain a target audio broadcast file in the audio format specified by the user; the present application not only helps to solve the personalized broadcast needs for local dialects or special languages ​​in mobile applications (APPs) through efficient voice merging and splicing and voice quality monitoring methods, but also can significantly reduce the high cost pressure caused by the frequent use of high-quality voice services, that is, to provide a simple implementation method for text-to-speech broadcast with excellent sound quality and conducive to saving enterprise operating costs.

[0089] In one embodiment, if Figure 2 As shown, in step S1, a database of multiple types of audio segment samples is pre-stored, and the multiple types of audio segment samples include language type, intonation type, speech speed type and timbre type, and each audio segment sample is associated with an audio feature identifier; based on the acquired voice broadcast setting instruction, multiple target audio segments to be processed and corresponding voice playback timing information are determined, including:

[0090] S11: Identify the corresponding text information to be broadcast and audio feature parameters based on the acquired voice broadcast setting instruction, and obtain the corresponding voice playback timing information based on the playback timing of the text information to be broadcast.

[0091] In this embodiment, when faced with the need for APP voice broadcast in a dialect other than Mandarin, in order to facilitate a simple implementation of APP voice broadcast, multiple types of audio clip samples are first pre-collected and stored in a database of an audio operating system (such as a mobile phone system API). The multiple types of audio clip samples include audio clip samples of different languages, different intonations, different speaking speeds and different timbres (as audio clip materials to be spliced ​​and merged), and the audio feature identifier is a unique identification code for the broadcast text information and audio feature parameter information of the audio clip sample.

[0092] S12: extracting multiple audio segment samples from the database based on the text information to be broadcast and the audio feature parameters according to the audio feature identifier of each audio segment sample; and determining the extracted multiple audio segment samples as target audio segments to be processed.

[0093] In this embodiment, the corresponding voice playback timing information is obtained based on the playback timing of the text information to be broadcasted according to the voice broadcast setting instruction set by the user. For example, if the text information to be broadcasted is the Cantonese voice and playback order of "Payment successful, amount 10.5 yuan", then in actual application, multiple audio segment samples of "Payment", "Success", "Amount", "10", "Point", "5" and "Yuan" in Cantonese can be extracted from the database as the target audio segment to be processed. At the same time, according to the method of extracting the audio feature identifier of each audio segment sample, it is beneficial to keep the audio feature parameters of the extracted multiple audio segment samples within a consistent unified numerical range; so as to facilitate the subsequent signal preprocessing and voice signal splicing and merging, thereby providing a simple and easy-to-implement APP voice broadcast method, and also providing users with voice interaction services that are more in line with actual needs.

[0094] In one embodiment, if Figure 3 As shown, in step S12, based on the text information to be broadcast and the audio feature parameters, multiple audio segment samples are extracted from the database according to the audio feature identifier of each audio segment sample, specifically including:

[0095] S121: In a database, audio segment samples whose text information to be broadcast matches the text information to be broadcast are screened to obtain a plurality of audio segment samples to be confirmed.

[0096] In this embodiment, the audio feature parameters include audio sampling rate, quantization bit depth and number of audio channels.

[0097] Specifically, when extracting multiple audio clip samples from the database, first preliminarily screen the multiple audio clip samples whose text information to be broadcast matches the text information to be broadcast (i.e., multiple audio clip samples to be confirmed), that is, preliminarily filter out the audio clip samples whose content information to be broadcast (i.e., text information) does not match.

[0098] S122: Identify and obtain set audio feature parameter values ​​of multiple audio segment samples to be confirmed according to the audio feature identifier of each audio segment sample.

[0099] Specifically, each audio segment sample has a one-to-one correspondence with an audio feature identifier; the set audio feature parameter values ​​of multiple audio segment samples to be confirmed are identified and obtained. Since the audio feature parameter values ​​of each audio segment sample material are set within a specified numerical range when the material of each audio segment sample is recorded in the early stage, the audio feature parameter values ​​of different audio segment samples are not the same.

[0100] S123: Determine an extraction parameter range of each audio feature parameter of each audio segment sample to be extracted based on the audio sampling rate, quantization bit depth and number of audio channels set in the voice broadcast setting instruction.

[0101] In this embodiment, the extraction parameter range of each audio feature parameter of each audio clip sample is the target parameter interval of each audio feature parameter of each audio clip sample, that is, the target parameter interval of the audio feature parameter at the audio sampling rate, quantization bit depth and number of audio channels.

[0102] S124: Based on the extraction parameter range of each audio feature parameter of each audio segment sample, extract the audio segment samples whose set audio feature parameter values ​​all meet the corresponding extraction parameter range.

[0103] In this embodiment, by confirming the requirements of the extraction parameter range of each target audio feature parameter to be extracted (referring to the audio sampling rate, quantization bit depth and number of audio channels), screening is performed according to the set audio feature parameter value of each audio feature parameter of each audio segment sample to obtain multiple audio segment samples whose audio feature parameters all meet the requirements of the voice broadcast setting instruction. At this time, the audio sampling rate, quantization bit depth and number of audio channels of the multiple audio segment samples screened out are consistent, that is, the audio sampling rate, quantization bit depth and number of audio channels among the multiple audio segment samples are at the same specified values, so that the language type, intonation type, speech speed type and timbre type among the multiple audio segment samples are consistent; this is conducive to ensuring that the spliced ​​audio can be played normally on the playback device with uniform sound quality, and the screening and extraction of audio segment samples is simple, fast and efficient.

[0104] In one embodiment, in step S2, audio preprocessing is performed on multiple target audio signals to set audio feature parameters of all target audio signals within a uniform value range, including:

[0105] S21: Perform volume intensity detection on multiple target audio signals to obtain corresponding volume intensities.

[0106] S22: When the volume intensity of a certain target audio signal exceeds a preset volume intensity threshold, normalize the target audio signal.

[0107] Specifically, a loop traversal method is used to detect the volume intensity of each target audio signal in turn, and when it is detected that the volume intensity of a certain target audio signal exceeds a preset volume intensity threshold, the target audio signal is normalized.

[0108] or,

[0109] S23: traverse and collect the audio tracks of multiple target audio clips during the broadcasting process to obtain corresponding sampling values.

[0110] In this embodiment, the audio tracks of the corresponding audio files of the plurality of target audio segments during the broadcasting process are traversed and collected to obtain the sampling values ​​of the plurality of audio volumes.

[0111] S24: Calculate the gain factor of each target audio signal according to the preset volume intensity threshold and the corresponding sampling value; multiply the audio track played by the target audio segment corresponding to all the target audio signals by the gain factor to obtain the corresponding adjusted audio track value, so as to perform gain processing on the target audio signal.

[0112] In this embodiment, the preset volume intensity threshold is the target volume value of the target audio segment at a certain sampling point, which can be pre-defined according to the actual APP voice broadcast requirements; thus, different sampling values ​​of the same target audio segment and the corresponding gain factors calculated by the volume intensity threshold are not the same.

[0113] Specifically, for target audio segments with inconsistent volumes, the present application provides a matching strategy for consistent volume matching by means of normalization processing or gain adjustment. Multiple target audio segments subjected to normalization processing and / or gain adjustment are beneficial for smooth transition of the audio segments after splicing and merging, thereby facilitating improving the quality of audio splicing.

[0114] In one embodiment, before step S3, the simple method for implementing APP voice broadcast further includes:

[0115] S31: According to a preset silent segment filtering rule, traverse and collect multiple target audio signals after normalization processing or gain processing to obtain multiple sampling point values.

[0116] In this embodiment, the sampling point value is the volume intensity value of the sampling point.

[0117] Specifically, through intelligent detection and recognition, the silent segments in each target audio signal (ie, each audio data stream) are automatically removed to improve the compactness and coherence of the voice broadcast.

[0118] S32: When a sampling point value of a target audio signal satisfies a preset silence segment filtering rule, the corresponding silence segment data stream of the target audio signal is filtered and removed; the preset silence segment filtering rule satisfies that the volume intensity values ​​of the sampling points of multiple consecutive frames are all lower than a preset silence volume threshold.

[0119] Specifically, a traversal method is used to detect whether there are silent segments in multiple target audio signals. That is, when the sampling point value of a certain target audio signal meets the filtering condition of the silent segment filtering rule, the sampling point value at this time is the volume intensity value of the sampling point, which represents that the signal data stream where the sampling point is located meets the volume intensity values ​​of the sampling points detected for multiple consecutive frames are all lower than the set silent volume threshold, thereby realizing the automatic removal of silent segments in each target audio signal. Only identification and filtering technology is used, which improves the compactness and continuity of APP voice broadcast.

[0120] In one embodiment, a simple method for implementing APP voice broadcast also includes:

[0121] S10: when the multiple target audio signals are spliced ​​together according to the splicing points to obtain the initial voice broadcast signal, the volume intensity of the splicing point in the initial voice broadcast signal is faded in according to a preset fade-in algorithm, and the volume intensity of the splicing point in the initial voice broadcast signal is faded out according to a preset fade-out algorithm;

[0122] The preset fade-in algorithm is shown in formula (1):

[0123] G(x)=Q(x)*e cx+b +δ*Y x (1)

[0124] Where G(x) is the volume intensity at time x, x is the time change of the preset time period for fade-in processing, Q(x) is the volume intensity of the splicing point, c and b are constants, δ is the adjustment coefficient, and Y x The calibrated volume intensity at time point x.

[0125] In this embodiment, the preset fade-out algorithm is shown in formula (2):

[0126] F(x)=Q(x)*e -(kx+d) +δ*Y x (2)

[0127] Where F(x) is the volume intensity at time x, x is the time change of the preset time period for fade-out processing, Q(x) is the volume intensity at the splicing point, k and d are constants, δ is the adjustment coefficient, and Y x The calibrated volume intensity at time point x.

[0128] Specifically, the target audio signal is a digitized audio data stream. By adopting a smooth transition splicing process of fade-in and fade-out processing, the transition between the target audio signals at both ends is natural and not abrupt. That is, the six target audio signal streams of "successful collection", "amount", "10", "points", "5" and "yuan" are smoothly spliced ​​together in sequence, ensuring that the broadcast content is logically clear and the transition is natural, giving users a listening experience like a real-time broadcast by a real person.

[0129] It should be understood that the serial numbers of the steps in the above embodiments do not imply a sequence of execution. The execution sequence of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0130] In one embodiment, a simple implementation system for APP voice broadcast is provided, and the simple implementation system for APP voice broadcast corresponds to a simple implementation method for APP voice broadcast in the above embodiment.

[0131] A simple implementation system for APP voice broadcasting includes an audio segment acquisition and decoding module, an audio signal preprocessing module, an audio segment splicing and merging module, an audio quality judgment and optimization module, and an audio file encoding module. The detailed description of each functional module is as follows:

[0132] The audio segment acquisition and decoding module is used to determine multiple target audio segments to be processed and corresponding voice playback timing information based on the acquired voice broadcast setting instruction; decode the multiple target audio segments to obtain corresponding multiple target audio signals;

[0133] An audio signal preprocessing module, used to perform audio preprocessing on multiple target audio signals, and set the audio feature parameters of all target audio signals within a unified value range;

[0134] The audio segment splicing and merging module is used to determine the splicing points between multiple target audio signals according to the voice playback timing information, and splice and merge the multiple target audio signals according to the splicing points to obtain an initial voice broadcast signal; the splicing point in the initial voice broadcast signal satisfies the connection between the end sampling point of the previous target audio signal and the start sampling point of the next target audio signal among the multiple target audio signals;

[0135] The audio quality judgment and optimization module is used to judge the voice quality of the initial voice broadcast signal according to the preset voice optimization judgment rules; when the initial voice broadcast signal meets the voice optimization processing conditions in the preset voice optimization judgment rules, the initial voice broadcast signal is optimized to obtain the target voice broadcast signal;

[0136] The audio file encoding module is used to encode the target voice broadcast signal to obtain a target audio broadcast file.

[0137] For the specific limitations of the simple implementation system of APP voice broadcast, please refer to the limitations of a simple implementation method of APP voice broadcast in the above text, which will not be repeated here; each module in the above-mentioned simple implementation system of APP voice broadcast can be implemented in whole or in part by software, hardware and their combination; the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0138] In one embodiment, a smart device is provided, which can be a server. The smart device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the smart device is used to provide computing and control capabilities. The memory of the smart device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the smart device is used to store a large amount of audio segment samples of various types, silent segment filtering rules, etc. The network interface of the smart device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a simple implementation method of an APP voice broadcast is implemented.

[0139] In one embodiment, a smart device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0140] S1: determining a plurality of target audio segments to be processed and corresponding voice playback timing information based on the acquired voice broadcast setting instruction; decoding the plurality of target audio segments to obtain a plurality of corresponding target audio signals;

[0141] S2: performing audio preprocessing on the multiple target audio signals, and setting the audio feature parameters of all the target audio signals within a uniform value range;

[0142] S3: determining the splicing points between the multiple target audio signals according to the voice playback timing information, splicing the multiple target audio signals according to the splicing points and obtaining an initial voice broadcast signal; the splicing points in the initial voice broadcast signal satisfy the connection between the end sampling point of the previous target audio signal in the multiple target audio signals and the start sampling point of the next target audio signal;

[0143] S4: judging the voice quality of the initial voice broadcast signal according to the preset voice optimization judgment rule; when the initial voice broadcast signal meets the voice optimization processing conditions in the preset voice optimization judgment rule, optimizing the initial voice broadcast signal to obtain the target voice broadcast signal;

[0144] S5: Encode the target voice broadcast signal to obtain a target audio broadcast file.

[0145] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0146] S1: determining a plurality of target audio segments to be processed and corresponding voice playback timing information based on the acquired voice broadcast setting instruction; decoding the plurality of target audio segments to obtain a plurality of corresponding target audio signals;

[0147] S2: performing audio preprocessing on the multiple target audio signals, and setting the audio feature parameters of all the target audio signals within a uniform value range;

[0148] S3: determining the splicing points between the multiple target audio signals according to the voice playback timing information, splicing the multiple target audio signals according to the splicing points and obtaining an initial voice broadcast signal; the splicing points in the initial voice broadcast signal satisfy the connection between the end sampling point of the previous target audio signal in the multiple target audio signals and the start sampling point of the next target audio signal;

[0149] S4: judging the voice quality of the initial voice broadcast signal according to the preset voice optimization judgment rule; when the initial voice broadcast signal meets the voice optimization processing conditions in the preset voice optimization judgment rule, optimizing the initial voice broadcast signal to obtain the target voice broadcast signal;

[0150] S5: Encode the target voice broadcast signal to obtain a target audio broadcast file.

[0151] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0152] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0153] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, it should be understood by those skilled in the art that the technical solutions described in the aforementioned embodiments may still be modified, or some of the features thereof may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A simple method for implementing APP voice broadcast, characterized in that: include: Determine multiple target audio segments to be processed and corresponding voice playback timing information based on the acquired voice broadcast setting instruction; decode the multiple target audio segments to obtain corresponding multiple target audio signals; Performing audio preprocessing on the plurality of target audio signals, and setting audio feature parameters of all the target audio signals within a uniform value range; the audio feature parameters include audio sampling rate, quantization bit depth, and number of audio channels; Determine the splicing points between the multiple target audio signals according to the voice playback timing information, splice the multiple target audio signals according to the splicing points to obtain an initial voice broadcast signal; the splicing point in the initial voice broadcast signal satisfies the connection between the end sampling point of the previous target audio signal and the start sampling point of the next target audio signal among the multiple target audio signals; The initial voice broadcast signal is subjected to a voice quality judgment according to a preset voice optimization judgment rule; when the initial voice broadcast signal satisfies the voice optimization processing condition in the preset voice optimization judgment rule, the initial voice broadcast signal is optimized to obtain a target voice broadcast signal; The target voice broadcast signal is encoded to obtain a target audio broadcast file.

2. According to claim 1, a simple method for implementing APP voice broadcast is characterized in that: A database pre-stored with multiple types of audio clip samples, the multiple types of audio clip samples include language type, intonation type, speech speed type and timbre type, and each audio clip sample is associated with an audio feature identifier; The method of determining a plurality of target audio segments to be processed and corresponding voice play timing information based on the acquired voice broadcast setting instruction includes: Based on the acquired voice broadcast setting instruction, the corresponding text information to be broadcast and the audio characteristic parameters are identified, and based on the playback timing of the text information to be broadcast, the corresponding voice playback timing information is obtained; Based on the text information to be broadcast and the audio feature parameters, multiple audio segment samples are extracted from the database according to the audio feature identifier of each audio segment sample; and the extracted multiple audio segment samples are determined as target audio segments to be processed.

3. According to claim 2, a simple method for implementing APP voice broadcast is characterized in that: The extracting a plurality of audio segment samples from the database based on the to-be-broadcasted text information and the audio feature parameters according to the audio feature identifier of each audio segment sample specifically includes: In the database, the audio segment samples whose text information to be broadcast matches the text information to be broadcast are screened to obtain a plurality of audio segment samples to be confirmed; According to the audio feature identifier of each audio segment sample, identifying and obtaining the set audio feature parameter values ​​of the plurality of audio segment samples to be confirmed; Determine an extraction parameter range of each audio feature parameter of each audio segment sample to be extracted based on the audio sampling rate, quantization bit depth and number of audio channels set in the voice broadcast setting instruction; Based on the extraction parameter range of each audio feature parameter of each audio segment sample, audio segment samples whose set audio feature parameter values ​​all conform to the corresponding extraction parameter range are extracted respectively.

4. According to claim 2, a simple method for implementing APP voice broadcast is characterized in that: The step of performing audio preprocessing on the plurality of target audio signals to set the audio feature parameters of all the target audio signals within a uniform value range includes: Performing volume intensity detection on the plurality of target audio signals to obtain corresponding volume intensities; When the volume intensity of a target audio signal exceeds a preset volume intensity threshold, normalizing the target audio signal; or, Traversing and collecting the audio tracks of the target audio segments during the broadcasting process to obtain corresponding sampling values; The gain factor of each target audio signal is calculated according to the preset volume intensity threshold and the corresponding sampling value; the audio track played by the target audio segment corresponding to all the target audio signals is multiplied by the gain factor to obtain the corresponding adjusted audio track value, so as to perform gain processing on the target audio signal.

5. According to claim 4, a simple method for implementing APP voice broadcast is characterized in that: Also includes: According to a preset silent segment filtering rule, multiple target audio signals after normalization or gain processing are traversed and collected to obtain multiple sampling point values; When the sampling point value of a target audio signal meets the preset silence segment filtering rule, the corresponding silence segment data stream of the target audio signal is filtered and removed; the preset silence segment filtering rule satisfies that the volume intensity values ​​of the sampling points of multiple consecutive frames are all lower than the preset silence volume threshold.

6. According to claim 1, a simple method for implementing APP voice broadcast is characterized in that: Also includes: When the plurality of target audio signals are spliced ​​together according to the splicing point to obtain an initial voice broadcast signal, a volume intensity of the splicing point in the initial voice broadcast signal is faded in according to a preset fade-in algorithm, and a volume intensity of the splicing point in the initial voice broadcast signal is faded out according to a preset fade-out algorithm; The preset fade-in algorithm is shown in formula (1): G(x)=Q(x)*e cx+b +δ*Y x (1) Where G(x) is the volume intensity at time x, x is the time change of the preset time period for fade-in processing, Q(x) is the volume intensity of the splicing point, c and b are constants, δ is the adjustment coefficient, and Y x The calibrated volume intensity at time point x.

7. According to claim 6, a simple method for implementing APP voice broadcast is characterized in that: The preset fade-out algorithm is shown in formula (2): F(x)=Q(x)*e -(kx+d) +δ*Y x (2) Where F(x) is the volume intensity at time x, x is the time change of the preset time period for fade-out processing, Q(x) is the volume intensity at the splicing point, k and d are constants, δ is the adjustment coefficient, and Y x The calibrated volume intensity at time point x.

8. A simple implementation system for APP voice broadcast, characterized in that: The system includes: The audio segment acquisition and decoding module is used to determine a plurality of target audio segments to be processed and corresponding voice playback timing information based on the acquired voice broadcast setting instruction; decode the plurality of target audio segments to obtain a plurality of corresponding target audio signals; An audio signal preprocessing module, used to perform audio preprocessing on the plurality of target audio signals, and set the audio feature parameters of all the target audio signals within a uniform value range; the audio feature parameters include audio sampling rate, quantization bit depth and number of audio channels; An audio segment splicing and merging module is used to determine the splicing points between the multiple target audio signals according to the voice playback timing information, and splice and merge the multiple target audio signals according to the splicing points to obtain an initial voice broadcast signal; the splicing point in the initial voice broadcast signal satisfies the connection between the end sampling point of the previous target audio signal and the starting sampling point of the next target audio signal among the multiple target audio signals; The audio quality judgment and optimization module is used to judge the broadcast voice quality of the initial voice broadcast signal according to the preset voice optimization judgment rule; when the initial voice broadcast signal meets the voice optimization processing conditions in the preset voice optimization judgment rule, the initial voice broadcast signal is optimized to obtain the target voice broadcast signal; The audio file encoding module is used to encode the target voice broadcast signal to obtain a target audio broadcast file.

9. An intelligent device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of a simple method for implementing APP voice broadcast as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by the processor, the steps of the simple method for implementing APP voice broadcast as described in any one of claims 1 to 7 are implemented.

Images