Multi-dimensional speech adaptive training method and system

Abstract

The invention discloses a multi-dimensional speech adaptive training method and system. The method comprises the following steps: collecting historical training data and training time intervals of a user, and judging whether vowel training is started or not; acquiring real-time training data of a user, and sequentially performing loudness judgment, pitch exercise judgment and vowel space area standard judgment; and respectively calculating corresponding vowel space area target adjustment coefficients according to the age and gender classification of the user, and entering different training formulating modes based on the size comparison of the second target and the first target to generate a personalized vowel exercise plan. Through dynamic acquisition and quantitative analysis of multi-dimensional voice parameters, self-adaptive adjustment of training content and difficulty is realized, manual fine adjustment is supported, the problems of parameter solidification, insufficient personalization and weak feedback mechanism in an existing voice training system are effectively solved, and the pertinence, adaptability and rehabilitation effect of training are improved.

Description

Technical Field

[0001] This invention relates to a method for multi-dimensional adaptive speech training, and also to a system for implementing this method, belonging to the field of speech training technology. Background Technology

[0002] Speech training technology has wide applications in language rehabilitation, pronunciation correction, and language learning. Its core objective is to improve users' pronunciation accuracy, stability, and speech control through systematic training. However, most existing speech training systems employ a fixed design, with parameters such as training content, difficulty, and intensity typically preset manually, lacking the ability to dynamically respond to individual user differences. For example, in the rehabilitation process of stroke-related aphasia patients, the types of speech impairments and rehabilitation stages vary significantly among different patients. A uniformly set training program is difficult to accurately adapt, resulting in insufficient training targeting and limited rehabilitation efficiency.

[0003] Furthermore, existing systems generally lack dynamic adjustment mechanisms based on real-time data during training. Although some systems can collect user speech features, they are mostly limited to simple accuracy statistics and fail to deeply analyze multi-dimensional speech features such as vowel space area, formant parameters, and fundamental frequency, nor have they established a data-driven automatic optimization link for training parameters. In addition, the training feedback is monotonous and fails to intuitively reflect pronunciation problems, making it difficult for users to adjust their pronunciation strategies accordingly, thus affecting the sustained participation and effectiveness of training.

[0004] The limitations of existing systems are particularly pronounced in scenarios that heavily rely on personalized progress, such as rehabilitation training and language learning. For example, in Chinese speech training, elements such as tone and rhythm are crucial for semantic expression, but traditional systems often use fixed training materials and cannot dynamically strengthen a user's weak points in pronunciation. Furthermore, these systems generally do not support users or therapists manually fine-tuning training parameters according to actual situations, limiting their flexibility and practicality in real-world applications. Summary of the Invention

[0005] The primary technical problem to be solved by this invention is to provide a method for multi-dimensional adaptive speech training.

[0006] Another technical problem to be solved by the present invention is to provide a system for multi-dimensional speech adaptive training.

[0007] To achieve the above-mentioned technical objectives, the present invention adopts the following technical solution: According to a first aspect of the present invention, a method for multi-dimensional speech adaptive training is provided, comprising the following steps: S1: Collect the user's historical training data and determine whether to enter training mode; S2: Collect user training time interval data to determine whether to enter vowel practice mode; S3: The user begins vowel practice, and user training data is collected; S4: Based on the user training data, determine whether the loudness is greater than or equal to the preset threshold; S5: Determine whether the training plan includes pitch practice; S6: Based on the training data, determine whether the vowel space area within the preset period meets the lower limit standard of the first vowel space area target; S7: Determine the user's age; S8: Determine the user's gender; S9: Calculate the adjustment coefficient of the vowel space area target for children and determine whether the second vowel space area target is smaller than the first vowel space area target; If yes, proceed to the first children's vowel practice program; if not, proceed to the second children's vowel practice program. S10: Calculate the adjustment coefficient of the male vowel space area target and determine whether the second vowel space area target is smaller than the first vowel space area target. If yes, proceed to the first male vowel practice program; if not, proceed to the second male vowel practice program. S11: Calculate the adjustment coefficient of the vowel space area target for females and determine whether the second vowel space area target is smaller than the first vowel space area target. If yes, proceed to the first female vowel practice program; if not, proceed to the second female vowel practice program. S12: Based on the results of S9 to S11, output the vowel practice plan and end the process.

[0008] Preferably, the user training data in step S3 includes: loudness, fundamental frequency, mandibular distance, tongue distance, vowel spatial area parameters, first formant of the centroid of vowel a, first formant of the centroid of vowel i, second formant of the centroid of vowel i, and second formant of the centroid of vowel u.

[0009] Preferably, in step S4, if the loudness is greater than or equal to a preset threshold, then proceed to step S5; if the loudness is less than the preset threshold, then determine whether the training plan includes loudness practice. If the training plan does not include loudness practice, proceed to step S5; if the training plan includes loudness practice, adjust the loudness practice parameters and proceed to step S5.

[0010] Preferably, the pitch practice determination in step S5 includes the following steps: First, determine whether the user's age is greater than a preset threshold; If the value is greater than the specified value, then gender determination will be performed. If the person is male, determine whether his fundamental frequency meets the preset first fundamental frequency standard; if yes, proceed to step S6; if not, adjust the pitch practice parameters and proceed to step S6. If the user is female, or if the user's age is less than or equal to a preset threshold, it is determined whether their fundamental frequency meets the preset second fundamental frequency standard; if yes, proceed to step S6; if not, conduct pitch practice until their fundamental frequency meets the preset second fundamental frequency standard, and then proceed to step S6.

[0011] Preferably, the first fundamental frequency standard is greater than 90 Hz and less than 300 Hz; the second fundamental frequency standard is greater than 140 Hz and less than 430 Hz.

[0012] Preferably, in step S6, the formula for calculating the lower limit standard of the training target is: Training target lower limit standard = first vowel space area target × first sensitivity coefficient; The formula for calculating the upper limit standard of the training target is: Upper limit of training target = target area of ​​first vowel space × second sensitivity coefficient; The formula for setting the area target of the second vowel space in the first type is: Second vowel space area target = First vowel space area target × Third sensitivity coefficient; The formula for setting the area target of the second vowel space is as follows: Second vowel space area target = First vowel space area target × Fourth sensitivity coefficient.

[0013] Preferably, in step S9, the formula for calculating the target adjustment coefficient for the vowel space area of ​​the child is: Adjustment factor for children's vowel space area target = target area of ​​second vowel space / children's adjustment factor; The first model for developing vowel exercises for children includes the following steps: Determine whether the mandibular distance is less than the child's vowel space area target adjustment factor × child's mandibular distance parameter; If not, determine whether the tongue distance is less than the child's vowel space area target adjustment coefficient × the child's tongue distance parameter; if yes, determine whether the child's vowel i first formant adjustment value is less than the vowel i centroid first formant. The formula for calculating the adjustment value of the first formant of the vowel i in children is as follows: Adjustment value of the first formant of vowel i in children = First parameter of vowel i in children - (Target adjustment coefficient of vowel space area in children × Mandibular distance parameter in children); If the adjustment value of the first formant of the child's vowel i is less than the first formant of the centroid of vowel i, the number of training sessions for vowel i will be increased according to the preset threshold; if the adjustment value of the first formant of the child's vowel i is greater than or equal to the first formant of the centroid of vowel i, it will be determined whether the adjustment value of the first formant of the child's vowel a is greater than the first formant of the centroid of vowel a. The formula for calculating the adjustment value of the first formant of the child vowel 'a' is as follows: Adjustment value of the first formant of the child's vowel a = parameter of the child's vowel a + (target adjustment coefficient of the child's vowel space area × parameter of the child's mandibular distance); If the adjustment value of the first formant of the child's vowel 'a' is greater than the first formant of the centroid of vowel 'a', then the number of training sessions for vowel 'a' will be increased according to the preset threshold; if the adjustment value of the first formant of the child's vowel 'a' is less than or equal to the first formant of the centroid of vowel 'a', then it will be determined whether the tongue distance is less than the child's target adjustment coefficient for vowel space area × the child's tongue distance parameter. If not, the number of training sessions for vowels o and e is reduced according to the preset threshold, and the process proceeds to step S12; if yes, it is determined whether the adjustment value of the second formant of the child's vowel i is greater than the second formant of the centroid of vowel i. The formula for calculating the adjustment value of the second formant of the vowel i in children is as follows: Adjustment value of the second formant of vowel i in children = Second parameter of vowel i in children + (Target adjustment coefficient of vowel space area in children × Tongue distance parameter in children); If the adjustment value of the second formant of the child's vowel i is greater than the second formant of the centroid of vowel i, then the number of training sessions for vowel i is increased according to the preset threshold; if the adjustment value of the second formant of the child's vowel i is less than or equal to the second formant of the centroid of vowel i, then it is determined whether the adjustment value of the second formant of the child's vowel u is less than the second formant of the centroid of vowel u. The formula for calculating the adjustment value of the second formant of the child vowel u is as follows: Adjustment value of the second formant of the vowel u in children = parameter of vowel u in children - (target adjustment coefficient of vowel space area in children × parameter of tongue distance in children); If the adjustment value of the second formant of the child's vowel u is less than the second formant of the centroid of the vowel u, then the number of training sessions for the vowel u is increased according to a preset threshold, and the number of training sessions for the vowels o and e is decreased according to a preset threshold, proceeding to step S12; if the adjustment value of the second formant of the child's vowel u is greater than or equal to the second formant of the centroid of the vowel u, then the number of training sessions for the vowels o and e is decreased according to a preset threshold, proceeding to step S12. In the second children's vowel practice development mode, the target area of ​​the second vowel space is greater than or equal to the area of ​​the first vowel space. The corresponding training content is: the number of training sessions for vowels i and u is reduced according to a preset threshold, and the number of training sessions for vowels a, o, and e is increased according to a preset threshold, and then proceed to step S12.

[0014] Preferably, in step S10, the formula for calculating the target adjustment coefficient for the vowel space area of ​​males is: Male vowel space area target adjustment factor = second vowel space area target / male adjustment factor; The first model for developing male vowel exercises includes the following steps: Determine whether the mandibular distance is less than the male vowel space area target adjustment coefficient × male mandibular distance parameter; If not, determine whether the tongue distance is less than the male vowel space area target adjustment coefficient × male tongue distance parameter; if yes, determine whether the adjustment value of the first formant of male vowel i is less than the first formant of the centroid of vowel i. The formula for calculating the adjustment value of the first formant of the male vowel i is as follows: Adjustment value of the first formant of male vowel i = First parameter of male vowel i - (Target adjustment coefficient of male vowel space area × Male mandibular distance parameter); If the adjustment value of the first formant of the male vowel i is less than the first formant of the centroid of vowel i, the number of training times for vowel i is increased according to the preset threshold; if the adjustment value of the first formant of the male vowel i is greater than or equal to the first formant of the centroid of vowel i, it is determined whether the adjustment value of the first formant of the male vowel a is greater than the first formant of the centroid of vowel a. The formula for calculating the adjustment value of the first formant of the male vowel 'a' is as follows: The adjustment value of the first formant of male vowel a = male vowel a parameter + (male vowel space area target adjustment coefficient × male mandibular distance parameter); If the adjustment value of the first formant of the male vowel a is greater than the first formant of the centroid of vowel a, then the number of training sessions for vowel a will be increased according to the preset threshold; if the adjustment value of the first formant of the male vowel a is less than or equal to the first formant of the centroid of vowel a, then it is determined whether the tongue distance is less than the male vowel space area target adjustment coefficient × male tongue distance parameter. If not, the number of training sessions for vowels o and e is reduced according to the preset threshold, and the process proceeds to step S12; if yes, it is determined whether the adjustment value of the second formant of the male vowel i is greater than the second formant of the centroid of vowel i. The formula for calculating the adjustment value of the second formant of the male vowel i is as follows: The adjustment value of the second formant of male vowel i = the second parameter of male vowel i + (the target adjustment coefficient of male vowel space area × male tongue distance parameter); If the adjustment value of the second formant of the male vowel i is greater than the second formant of the centroid of vowel i, the number of training times for vowel i is increased according to the preset threshold; if the adjustment value of the second formant of the male vowel i is less than or equal to the second formant of the centroid of vowel i, it is determined whether the adjustment value of the second formant of the male vowel u is less than the second formant of the centroid of vowel u. The formula for calculating the adjustment value of the second formant of the male vowel u is as follows: Adjustment value of the second formant of male vowel u = male vowel u parameter - (male vowel space area target adjustment coefficient × male tongue distance parameter); If the adjustment value of the second formant of the male vowel u is less than the second formant of the centroid of vowel u, then the number of training sessions for vowel u is increased according to a preset threshold, and the number of training sessions for vowels o and e is decreased according to a preset threshold, and proceed to step S12; if the adjustment value of the second formant of the male vowel u is greater than or equal to the second formant of the centroid of vowel u, then the number of training sessions for vowels o and e is decreased according to a preset threshold, and proceed to step S12. In the second male vowel practice design mode, the target area of ​​the second vowel space is greater than or equal to the area of ​​the first vowel space. The corresponding training content is as follows: the number of training sessions for vowels i and u is reduced according to a preset threshold, and the number of training sessions for vowels a, o, and e is increased according to a preset threshold, and then proceed to step S12.

[0015] Preferably, in step S11, the formula for calculating the target adjustment coefficient for the vowel space area of ​​females is: Female vowel space area target adjustment factor = second vowel space area target / female adjustment factor; The first model for developing female vowel exercises includes the following steps: Determine whether the mandibular distance is less than the female vowel space area target adjustment coefficient × female mandibular distance parameter; If not, determine whether the tongue distance is less than the female vowel space area target adjustment coefficient × female tongue distance parameter; if yes, determine whether the female vowel i first formant adjustment value is less than the vowel i centroid first formant. The formula for calculating the adjustment value of the first formant of the female vowel i is as follows: The adjustment value of the first formant of the female vowel i = the first parameter of the female vowel i - (the target adjustment coefficient of the female vowel space area × the female mandibular distance parameter); If the adjustment value of the first formant of the female vowel i is less than the first formant of the centroid of vowel i, the number of training times for vowel i is increased according to the preset threshold; if the adjustment value of the first formant of the female vowel i is greater than or equal to the first formant of the centroid of vowel i, it is determined whether the adjustment value of the first formant of the female vowel a is greater than the first formant of the centroid of vowel a. The formula for calculating the adjustment value of the first formant of the female vowel 'a' is as follows: The adjustment value of the first formant of the female vowel a = the parameter of the female vowel a + (the target adjustment coefficient of the female vowel space area × the female mandibular distance parameter). If the adjustment value of the first formant of the female vowel 'a' is greater than the first formant of the centroid of vowel 'a', then the number of training sessions for vowel 'a' will be increased according to the preset threshold; if the adjustment value of the first formant of the female vowel 'a' is less than or equal to the first formant of the centroid of vowel 'a', then it will be determined whether the tongue distance is less than the female vowel space area target adjustment coefficient × female tongue distance parameter. If not, the number of training sessions for vowels o and e is reduced according to the preset threshold, and the process proceeds to step S12; if yes, it is determined whether the adjustment value of the second formant of the female vowel i is greater than the second formant of the centroid of vowel i. The formula for calculating the adjustment value of the second formant of the female vowel i is as follows: The adjustment value of the second formant of the female vowel i = the second parameter of the female vowel i + (the target adjustment coefficient of the female vowel space area × the female tongue distance parameter); If the adjustment value of the second formant of the female vowel i is greater than the second formant of the centroid of vowel i, the number of training times for vowel i is increased according to the preset threshold; if the adjustment value of the second formant of the female vowel i is less than or equal to the second formant of the centroid of vowel i, it is determined whether the adjustment value of the second formant of the female vowel u is less than the second formant of the centroid of vowel u. The formula for calculating the adjustment value of the second formant of the female vowel u is as follows: The adjustment value of the second formant of the female vowel u = the female vowel u parameter - (the female vowel space area target adjustment coefficient × the female tongue distance parameter); If the adjustment value of the second formant of the female vowel u is less than the second formant of the centroid of vowel u, then the number of training sessions for vowel u is increased according to a preset threshold, and the number of training sessions for vowels o and e is decreased according to a preset threshold, and proceed to step S12; if the adjustment value of the second formant of the female vowel u is greater than or equal to the second formant of the centroid of vowel u, then the number of training sessions for vowels o and e is decreased according to a preset threshold, and proceed to step S12. In the second female vowel practice design mode, the target area of ​​the second vowel space is greater than or equal to the area of ​​the first vowel space. The corresponding training content is as follows: the number of training sessions for vowels i and u is reduced according to a preset threshold, and the number of training sessions for vowels a, o, and e is increased according to a preset threshold, and then proceed to step S12.

[0016] According to a second aspect of the present invention, a system for multi-dimensional voice adaptive training is provided, including a processor and a memory; wherein the memory is coupled to the processor and is used to store a computer program, which, when executed by the processor, causes the processor to implement the above-described method.

[0017] Compared with existing technologies, this invention, by collecting multi-dimensional speech feature data from users in real time and combining it with age and gender differences, constructs a dynamic evaluation system based on multiple indicators such as vowel space area, formant parameters, fundamental frequency, and loudness. This enables the quantitative and automatic adjustment of training difficulty, intensity, and content. Using this invention, not only can personalized training plans be adaptively generated based on the user's real-time performance, but it also supports manual parameter fine-tuning by users or therapists. This effectively solves the problems of fixed parameters, insufficient personalization, lack of dynamic feedback, and lack of human-computer collaboration capabilities in existing speech training systems, significantly improving the relevance, adaptability, and training efficiency of speech training in rehabilitation therapy and language learning scenarios. Attached Figure Description

[0018] Figure 1 A flowchart of a multi-dimensional speech adaptive training method provided in the first embodiment of the present invention; Figure 2 This is a flowchart of pitch practice determination in the first embodiment of the present invention; Figure 3 This is a schematic diagram of a multi-dimensional speech adaptive training system provided in the second embodiment of the present invention. Detailed Implementation

[0019] The technical content of the present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0020] First Embodiment like Figure 1 As shown, the first embodiment of the present invention provides a method for multi-dimensional speech adaptive training, which includes at least the following steps: S1: Collect the user's historical training data and determine whether to enter training mode.

[0021] If the user's cumulative training time is greater than or equal to the preset training time within the preset time period, the process ends; if the user's cumulative training time is less than the preset training time within the preset time period, proceed to step S2.

[0022] S2: Collect user training time interval data to determine whether to enter vowel practice mode.

[0023] If the time interval between the user's last training session and the last training session is greater than a preset threshold, proceed to step S3; if the time interval between the user's last training session and the last training session is less than or equal to the preset threshold, end the process.

[0024] S3: The user begins vowel practice, and user training data is collected.

[0025] S4: Based on the user's training data, determine whether the loudness is greater than or equal to the preset threshold.

[0026] If the loudness is greater than or equal to the preset threshold, proceed to step S5; if the loudness is less than the preset threshold, determine whether the training plan includes loudness practice.

[0027] If the training plan does not include loudness practice, proceed to step S5; if the training plan includes loudness practice, adjust the loudness practice parameters and proceed to step S5.

[0028] S5: Determine whether the training plan includes pitch practice.

[0029] If the training plan does not include pitch practice, proceed to step S6; if the training plan includes pitch practice, proceed to the pitch practice determination step.

[0030] S6: Based on the training data, determine whether the vowel space area within the preset period meets the lower limit standard of the first vowel space area target.

[0031] If the vowel space area within the preset period does not meet the lower limit standard of the first vowel space area target, then the second vowel space area target is formulated using the first calculation method, and proceed to step S7; if the vowel space area within the preset period meets the lower limit standard of the first vowel space area target, then it is determined whether the vowel space area within the preset period meets the upper limit standard of the first vowel space area target. If the vowel space area within the preset period does not meet the upper limit standard of the first vowel space area target, the process ends; if the vowel space area within the preset period meets the upper limit standard of the first vowel space area target, the second vowel space area target is determined using the second calculation method, and the process proceeds to step S7.

[0032] S7: Determines the user's age.

[0033] If the user's age is greater than the preset age threshold, proceed to step S8; if the user's age is less than or equal to the preset age threshold, proceed to step S9.

[0034] S8: Determine the user's gender.

[0035] If the user is male, proceed to step S10; if the user is female, proceed to step S11.

[0036] S9: Calculate the adjustment coefficient of the vowel space area target for children and determine whether the second vowel space area target is smaller than the first vowel space area target.

[0037] If yes, proceed to the first children's vowel practice program; if not, proceed to the second children's vowel practice program.

[0038] S10: Calculate the adjustment coefficient of the male vowel space area target and determine whether the second vowel space area target is smaller than the first vowel space area target.

[0039] If yes, proceed to the first male vowel practice program; if not, proceed to the second male vowel practice program.

[0040] S11: Calculate the adjustment coefficient of the vowel space area target for females and determine whether the second vowel space area target is smaller than the first vowel space area target.

[0041] If yes, proceed to the first female vowel practice program; if not, proceed to the second female vowel practice program.

[0042] S12: Based on the results of S9 to S11, output the vowel practice plan and end the process.

[0043] In one embodiment of the present invention, the preset time period in step S1 is preferably 5 days, and the preferred cumulative training time for users is 25 minutes. In step S2, the preferred time interval is 5 days. The user training data in step S3 includes: loudness, fundamental frequency, mandibular distance, tongue distance, vowel space area (VSA) parameter, first formant of the centroid of vowel a, first formant of the centroid of vowel i, second formant of the centroid of vowel i, and second formant of the centroid of vowel u.

[0044] In a preferred embodiment of the present invention, the loudness threshold in step S4 is 50 dBA.

[0045] like Figure 2 As shown, in one embodiment of the present invention, the pitch practice determination in step S5 includes the following steps: First, determine whether the user's age is greater than a preset threshold.

[0046] If the value is greater than the specified value, then gender determination will be performed. If the person is male, determine whether his fundamental frequency meets the preset first fundamental frequency standard; if yes, proceed to step S6; if not, adjust the pitch practice parameters and proceed to step S6. If the user is female, or if the user's age is less than or equal to a preset threshold, then determine whether their fundamental frequency meets the preset second fundamental frequency standard; if yes, proceed to step S6; if not, adjust the pitch practice parameters and proceed to step S6.

[0047] Preferably, the preset threshold for user age is 14 years old.

[0048] In a preferred embodiment of the present invention, the first fundamental frequency standard is greater than 90 Hz and less than 300 Hz; the second fundamental frequency standard is greater than 140 Hz and less than 430 Hz.

[0049] Preferably, based on the training data within a preset period, it is determined whether the training content needs to be optimized.

[0050] The judgment conditions include: 1. Training duration exceeds 6 minutes.

[0051] 2. The training period exceeds 5 days, and more than 5 days have passed since the last adjustment.

[0052] 3. Pronunciation accuracy is less than 60% or greater than 90%.

[0053] Based on the judgment criteria, the following adjustments will be made: 1. If the pronunciation accuracy is less than 60%, and the current pronunciation precision is less than the maximum possible pronunciation precision, then the pronunciation precision will be increased by 10 cents.

[0054] 2. If the pronunciation accuracy is greater than 90%, and the current pronunciation precision is greater than the minimum possible pronunciation precision, then the pronunciation precision will be reduced by 10 cents.

[0055] 3. If the training duration is greater than 10 minutes and the current training intensity is greater than the minimum set value, the training intensity will be reduced by 60%.

[0056] 4. If the training duration is less than 5 minutes and the current training intensity is less than the maximum set value, the training intensity will be increased by 60.

[0057] In one embodiment of the present invention, the formula for calculating the lower limit standard of the training target in step S6 is as follows: Training target lower limit standard = first vowel space area target × first sensitivity coefficient Preferably, the first sensitivity coefficient is 0.65.

[0058] The formula for calculating the upper limit standard of the training target is: Upper limit of training target = target area of ​​first vowel space × second sensitivity coefficient Preferably, the second sensitivity coefficient is 1.6.

[0059] The formula for setting the area target of the second vowel space in the first type is: Second vowel space area target = First vowel space area target × Third sensitivity coefficient Preferably, the third sensitivity coefficient is 0.6.

[0060] The formula for setting the area target of the second vowel space is as follows: Second vowel space area target = First vowel space area target × Fourth sensitivity coefficient Preferably, the fourth sensitivity coefficient is 1.3.

[0061] In one embodiment of the present invention, preferably, the preset age threshold in step S7 is 14 years old.

[0062] In one embodiment of the present invention, in step S9, the formula for calculating the child's vowel space area target adjustment coefficient is: Child's vowel space area target adjustment coefficient = Second vowel space area target / Child's adjustment coefficient In a preferred embodiment of the present invention, the adjustment factor for children is 20.67.

[0063] The first model for developing vowel practice for children includes the following steps: Determine whether the mandibular distance is less than the child's vowel space area target adjustment factor × child's mandibular distance parameter.

[0064] In a preferred embodiment of the present invention, the mandibular distance parameter for children is 5.33.

[0065] If not, determine whether the tongue distance is less than the child's vowel space area target adjustment coefficient × the child's tongue distance parameter; if yes, determine whether the child's vowel i first formant adjustment value is less than the vowel i centroid first formant. The formula for calculating the adjustment value of the first formant of the vowel i in children is: Adjustment value of the first formant of the vowel i in children = First parameter of the vowel i in children - (Target adjustment coefficient of vowel space area in children × Mandibular distance parameter in children) In a preferred embodiment of the present invention, the first parameter of the child vowel i is 8.62.

[0066] If the adjustment value of the first formant of the child's vowel i is less than the first formant of the centroid of vowel i, the number of training sessions for vowel i will be increased according to the preset threshold; if the adjustment value of the first formant of the child's vowel i is greater than or equal to the first formant of the centroid of vowel i, it will be determined whether the adjustment value of the first formant of the child's vowel a is greater than the first formant of the centroid of vowel a. The formula for calculating the adjustment value of the first formant of the vowel 'a' in children is: Adjustment value of the first formant of the vowel 'a' in children = Parameter of the vowel 'a' in children + (Target adjustment coefficient of vowel space area in children × Parameter of mandibular distance in children) In a preferred embodiment of the present invention, the parameter of the child vowel 'a' is 3.29.

[0067] If the adjustment value of the first formant of the child's vowel 'a' is greater than the first formant of the centroid of vowel 'a', then the number of training sessions for vowel 'a' will be increased according to the preset threshold; if the adjustment value of the first formant of the child's vowel 'a' is less than or equal to the first formant of the centroid of vowel 'a', then it will be determined whether the tongue distance is less than the child's target adjustment coefficient for vowel space area × the child's tongue distance parameter.

[0068] In a preferred embodiment of the present invention, the tongue distance parameter for children is 8.23.

[0069] If not, the number of training sessions for vowels o and e is reduced according to the preset threshold, and the process proceeds to step S12; if yes, it is determined whether the adjustment value of the second formant of the child's vowel i is greater than the second formant of the centroid of vowel i. The formula for calculating the adjustment value of the second formant of the vowel i in children is: Adjustment value of the second formant of the vowel i in children = Second parameter of the vowel i in children + (Target adjustment coefficient of vowel space area in children × Parameter of tongue distance in children) In a preferred embodiment of the present invention, the second parameter of the child vowel i is 7.49.

[0070] If the adjustment value of the second formant of the child's vowel i is greater than the second formant of the centroid of vowel i, then the number of training sessions for vowel i is increased according to the preset threshold; if the adjustment value of the second formant of the child's vowel i is less than or equal to the second formant of the centroid of vowel i, then it is determined whether the adjustment value of the second formant of the child's vowel u is less than the second formant of the centroid of vowel u. The formula for calculating the adjustment value of the second formant of the vowel u in children is: Adjustment value of the second formant of the vowel u in children = Parameter of vowel u in children - (Target adjustment coefficient of vowel space area in children × Parameter of tongue distance in children) In a preferred embodiment of the present invention, the parameter of the child vowel u is 15.73.

[0071] If the adjustment value of the second formant of the child's vowel u is less than the second formant of the centroid of the vowel u, then the number of training sessions for the vowel u is increased according to a preset threshold, and the number of training sessions for the vowels o and e is decreased according to a preset threshold, proceeding to step S12; if the adjustment value of the second formant of the child's vowel u is greater than or equal to the second formant of the centroid of the vowel u, then the number of training sessions for the vowels o and e is decreased according to a preset threshold, proceeding to step S12.

[0072] In the second children's vowel practice development mode, the target area of ​​the second vowel space is greater than or equal to the area of ​​the first vowel space. The corresponding training content is: the number of training sessions for vowels i and u is reduced according to a preset threshold, and the number of training sessions for vowels a, o, and e is increased according to a preset threshold, and then proceed to step S12.

[0073] In one embodiment of the present invention, in step S10, the formula for calculating the male vowel space area target adjustment coefficient is: Male vowel space area target adjustment coefficient = Second vowel space area target / Male adjustment coefficient In a preferred embodiment of the present invention, the male adjustment factor is 11.36.

[0074] The first method for developing male vowel exercises includes the following steps: Determine whether the mandibular distance is less than the male vowel space area target adjustment factor × male mandibular distance parameter.

[0075] In a preferred embodiment of the present invention, the male mandibular distance parameter is 3.56.

[0076] If not, determine whether the tongue distance is less than the male vowel space area target adjustment coefficient × male tongue distance parameter; if yes, determine whether the adjustment value of the first formant of male vowel i is less than the first formant of the centroid of vowel i. The formula for calculating the adjustment value of the first formant of male vowel i is: Adjustment value of the first formant of male vowel i = First parameter of male vowel i - (Target adjustment coefficient of male vowel space area × Male mandibular distance parameter) In a preferred embodiment of the present invention, the first parameter of the male vowel i is 6.37.

[0077] If the adjustment value of the first formant of the male vowel i is less than the first formant of the centroid of vowel i, the number of training times for vowel i is increased according to the preset threshold; if the adjustment value of the first formant of the male vowel i is greater than or equal to the first formant of the centroid of vowel i, it is determined whether the adjustment value of the first formant of the male vowel a is greater than the first formant of the centroid of vowel a. The formula for calculating the adjustment value of the first formant of the male vowel 'a' is: Adjustment value of the first formant of the male vowel 'a' = Male vowel 'a' parameter + (Male vowel space area target adjustment coefficient × Male mandibular distance parameter) In a preferred embodiment of the present invention, the parameter of the male vowel 'a' is 2.81.

[0078] If the adjustment value of the first formant of the male vowel 'a' is greater than the first formant of the centroid of vowel 'a', then the number of training sessions for vowel 'a' is increased according to the preset threshold; if the adjustment value of the first formant of the male vowel 'a' is less than or equal to the first formant of the centroid of vowel 'a', then it is determined whether the tongue distance is less than the male vowel space area target adjustment coefficient × male tongue distance parameter.

[0079] In a preferred embodiment of the present invention, the male tongue distance parameter is 7.23.

[0080] If not, the number of training sessions for vowels o and e is reduced according to the preset threshold, and the process proceeds to step S12; if yes, it is determined whether the adjustment value of the second formant of the male vowel i is greater than the second formant of the centroid of vowel i. The formula for calculating the adjustment value of the second formant of male vowel i is: Adjustment value of the second formant of male vowel i = Second parameter of male vowel i + (Target adjustment coefficient of male vowel space area × Male tongue distance parameter). In a preferred embodiment of the present invention, the second parameter of the male vowel i is 6.53.

[0081] If the adjustment value of the second formant of the male vowel i is greater than the second formant of the centroid of vowel i, the number of training times for vowel i is increased according to the preset threshold; if the adjustment value of the second formant of the male vowel i is less than or equal to the second formant of the centroid of vowel i, it is determined whether the adjustment value of the second formant of the male vowel u is less than the second formant of the centroid of vowel u. The formula for calculating the adjustment value of the second formant of the male vowel u is: Adjustment value of the second formant of the male vowel u = Male vowel u parameter - (Male vowel space area target adjustment coefficient × Male tongue distance parameter) In a preferred embodiment of the present invention, the male vowel u parameter is 13.76.

[0082] If the adjustment value of the second formant of the male vowel u is less than the second formant of the centroid of vowel u, then the number of training sessions for vowel u is increased according to a preset threshold, and the number of training sessions for vowels o and e is decreased according to a preset threshold, proceeding to step S12; if the adjustment value of the second formant of the male vowel u is greater than or equal to the second formant of the centroid of vowel u, then the number of training sessions for vowels o and e is decreased according to a preset threshold, proceeding to step S12.

[0083] In the second male vowel practice design mode, the target area of ​​the second vowel space is greater than or equal to the area of ​​the first vowel space. The corresponding training content is as follows: the number of training sessions for vowels i and u is reduced according to a preset threshold, and the number of training sessions for vowels a, o, and e is increased according to a preset threshold, and then proceed to step S12.

[0084] In one embodiment of the present invention, in step S11, the formula for calculating the target adjustment coefficient of the vowel space area for females is: Target adjustment coefficient of the vowel space area for females = Second vowel space area target / Female adjustment coefficient In a preferred embodiment of the present invention, the female adjustment factor is 13.44.

[0085] The first model for developing female vowel exercises includes the following steps: Determine whether the mandibular distance is less than the female vowel space area target adjustment factor × female mandibular distance parameter.

[0086] In a preferred embodiment of the present invention, the distance parameter of the female mandible is 3.89.

[0087] If not, determine whether the tongue distance is less than the female vowel space area target adjustment coefficient × female tongue distance parameter; if yes, determine whether the female vowel i first formant adjustment value is less than the vowel i centroid first formant. The formula for calculating the adjustment value of the first formant of the female vowel i is: Adjustment value of the first formant of the female vowel i = First parameter of the female vowel i - (Target adjustment coefficient of the female vowel space area × Female mandibular distance parameter). In a preferred embodiment of the present invention, the first parameter of the female vowel i is 7.03.

[0088] If the adjustment value of the first formant of the female vowel i is less than the first formant of the centroid of vowel i, the number of training times for vowel i is increased according to the preset threshold; if the adjustment value of the first formant of the female vowel i is greater than or equal to the first formant of the centroid of vowel i, it is determined whether the adjustment value of the first formant of the female vowel a is greater than the first formant of the centroid of vowel a. The formula for calculating the adjustment value of the first formant of the female vowel 'a' is: Adjustment value of the first formant of the female vowel 'a' = Female vowel 'a' parameter + (Target adjustment coefficient for female vowel space area × Female mandibular distance parameter) In a preferred embodiment of the present invention, the parameter of the female vowel 'a' is 3.14.

[0089] If the adjustment value of the first formant of the female vowel 'a' is greater than the first formant of the centroid of vowel 'a', then the number of training sessions for vowel 'a' is increased according to the preset threshold; if the adjustment value of the first formant of the female vowel 'a' is less than or equal to the first formant of the centroid of vowel 'a', then it is determined whether the tongue distance is less than the female vowel space area target adjustment coefficient × female tongue distance parameter.

[0090] In a preferred embodiment of the present invention, the female tongue distance parameter is 8.51.

[0091] If not, the number of training sessions for vowels o and e is reduced according to the preset threshold, and the process proceeds to step S12; if yes, it is determined whether the adjustment value of the second formant of the female vowel i is greater than the second formant of the centroid of vowel i. The formula for calculating the adjustment value of the second formant of the female vowel i is: Adjustment value of the second formant of the female vowel i = Second parameter of the female vowel i + (Target adjustment coefficient of the female vowel space area × Female tongue distance parameter). In a preferred embodiment of the present invention, the second parameter of the female vowel i is 6.88.

[0092] If the adjustment value of the second formant of the female vowel i is greater than the second formant of the centroid of vowel i, the number of training times for vowel i is increased according to the preset threshold; if the adjustment value of the second formant of the female vowel i is less than or equal to the second formant of the centroid of vowel i, it is determined whether the adjustment value of the second formant of the female vowel u is less than the second formant of the centroid of vowel u. The formula for calculating the adjustment value of the second formant of the female vowel u is: Adjustment value of the second formant of the female vowel u = Female vowel u parameter - (Target adjustment coefficient of female vowel space area × Female tongue distance parameter) In a preferred embodiment of the present invention, the female vowel u parameter is 15.4.

[0093] If the adjustment value of the second formant of the female vowel u is less than the second formant of the centroid of vowel u, the number of training sessions for vowel u is increased according to a preset threshold, and the number of training sessions for vowels o and e is decreased according to a preset threshold, proceeding to step S12; if the adjustment value of the second formant of the female vowel u is greater than or equal to the second formant of the centroid of vowel u, the number of training sessions for vowels o and e is decreased according to a preset threshold, proceeding to step S12.

[0094] In the second female vowel practice design mode, the target area of ​​the second vowel space is greater than or equal to the area of ​​the first vowel space. The corresponding training content is as follows: the number of training sessions for vowels i and u is reduced according to a preset threshold, and the number of training sessions for vowels a, o, and e is increased according to a preset threshold, and then proceed to step S12.

[0095] Second Embodiment Based on the above method, the second embodiment of the present invention provides a system for multi-dimensional speech adaptive training. For example... Figure 3 As shown, the system includes one or more processors and a memory. The memory is coupled to the processor and is used to store computer programs that, when executed by the processor, can implement the methods described in the above embodiments.

[0096] The processor controls the overall operation of the system to complete all or part of the steps described above. This processor can be a central processing unit (CPU), graphics processing unit (GPU), field-programmable gate array (FPGA), application-specific integrated circuit (ASIC), digital signal processing (DSP) chip, etc. The memory stores various types of data to support the operation of the system. This data may include, for example, instructions for any application or method operating on the system, as well as application-related data. The memory can be implemented using any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, etc.

[0097] In one exemplary embodiment, the system may be implemented by a computer chip or physical entity, or by a product with certain functions, to perform the methods described above and achieve the same technical effects as those methods described above. A typical embodiment is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, an in-vehicle human-machine interface device, a cellular phone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or any combination of these devices.

[0098] In another exemplary embodiment, the present invention also provides a computer-readable storage medium including program instructions that, when executed by a processor, implement the steps of the method in any of the above embodiments. For example, the computer-readable storage medium may be the memory including the program instructions described above, which can be executed by a processor to complete the above method and achieve the same technical effects as the above method.

[0099] It should be noted that the above embodiments are merely illustrative examples, and the technical solutions of each embodiment can be combined, all of which are within the protection scope of this invention.

[0100] The multi-dimensional speech adaptive training method and system provided by this invention have been described in detail above. Any obvious modifications made by those skilled in the art without departing from the essence of this invention will constitute an infringement of the patent rights of this invention and will incur corresponding legal liability.

Claims

1. A method for multi-dimensional adaptive speech training, characterized in that... Includes the following steps: S1: Collect the user's historical training data and determine whether to enter training mode; S2: Collect user training time interval data to determine whether to enter vowel practice mode; S3: The user begins vowel practice, and user training data is collected; S4: Based on the user training data, determine whether the loudness is greater than or equal to the preset threshold; S5: Determine whether the training plan includes pitch practice; S6: Based on the training data, determine whether the vowel space area within the preset period meets the lower limit standard of the first vowel space area target; S7: Determine the user's age; S8: Determine the user's gender; S9: Calculate the adjustment coefficient of the vowel space area target for children and determine whether the second vowel space area target is smaller than the first vowel space area target; If yes, proceed to the first children's vowel practice program; if not, proceed to the second children's vowel practice program. S10: Calculate the adjustment coefficient of the male vowel space area target and determine whether the second vowel space area target is smaller than the first vowel space area target. If yes, proceed to the first male vowel practice program; if not, proceed to the second male vowel practice program. S11: Calculate the adjustment coefficient of the vowel space area target for females and determine whether the second vowel space area target is smaller than the first vowel space area target. If yes, proceed to the first female vowel practice program; if not, proceed to the second female vowel practice program. S12: Based on the results of S9 to S11, output the vowel practice plan and end the process.

2. The method as described in claim 1, characterized in that... The user training data in step S3 includes: loudness, fundamental frequency, mandibular distance, tongue distance, vowel space area parameters, first formant of the centroid of vowel a, first formant of the centroid of vowel i, second formant of the centroid of vowel i, and second formant of the centroid of vowel u.

3. The method as described in claim 1, characterized in that... In step S4, if the loudness is greater than or equal to a preset threshold, proceed to step S5; if the loudness is less than the preset threshold, determine whether the training plan includes loudness practice. If the training plan does not include loudness practice, proceed to step S5; if the training plan includes loudness practice, adjust the loudness practice parameters and proceed to step S5.

4. The method as described in claim 1, characterized in that... The pitch practice determination in step S5 includes the following steps: First, determine whether the user's age is greater than a preset threshold; If the value is greater than the specified value, then gender determination will be performed. If the person is male, determine whether his fundamental frequency meets the preset first fundamental frequency standard; if yes, proceed to step S6; if not, adjust the pitch practice parameters and proceed to step S6. If the user is female, or if the user's age is less than or equal to a preset threshold, it is determined whether their fundamental frequency meets the preset second fundamental frequency standard; if yes, proceed to step S6; if not, conduct pitch practice until their fundamental frequency meets the preset second fundamental frequency standard, and then proceed to step S6.

5. The method as described in claim 4, characterized in that... The first fundamental frequency standard is greater than 90 Hz and less than 300 Hz; the second fundamental frequency standard is greater than 140 Hz and less than 430 Hz.

6. The method as described in claim 1, characterized in that... The formula for calculating the lower limit standard of the training target in step S6 is as follows: Training target lower limit standard = first vowel space area target × first sensitivity coefficient; The formula for calculating the upper limit standard of the training target is: Upper limit of training target = target area of ​​first vowel space × second sensitivity coefficient; The formula for setting the area target of the second vowel space in the first type is: Second vowel space area target = First vowel space area target × Third sensitivity coefficient; The formula for setting the area target of the second vowel space is as follows: Second vowel space area target = First vowel space area target × Fourth sensitivity coefficient.

7. The method as described in claim 1, characterized in that... In step S9, the formula for calculating the target adjustment coefficient for the vowel space area of ​​the child is as follows: Adjustment factor for children's vowel space area target = target area of ​​second vowel space / children's adjustment factor; The first model for developing vowel exercises for children includes the following steps: Determine whether the mandibular distance is less than the child's vowel space area target adjustment factor × child's mandibular distance parameter; If not, determine whether the tongue distance is less than the child's vowel space area target adjustment coefficient × the child's tongue distance parameter; if yes, determine whether the child's vowel i first formant adjustment value is less than the vowel i centroid first formant. The formula for calculating the adjustment value of the first formant of the vowel i in children is as follows: Adjustment value of the first formant of vowel i in children = First parameter of vowel i in children - (Target adjustment coefficient of vowel space area in children × Mandibular distance parameter in children); If the adjustment value of the first formant of the child's vowel i is less than the first formant of the centroid of vowel i, the number of training sessions for vowel i will be increased according to the preset threshold; if the adjustment value of the first formant of the child's vowel i is greater than or equal to the first formant of the centroid of vowel i, it will be determined whether the adjustment value of the first formant of the child's vowel a is greater than the first formant of the centroid of vowel a. The formula for calculating the adjustment value of the first formant of the child vowel 'a' is as follows: Adjustment value of the first formant of the child's vowel a = parameter of the child's vowel a + (target adjustment coefficient of the child's vowel space area × parameter of the child's mandibular distance); If the adjustment value of the first formant of the child's vowel 'a' is greater than the first formant of the centroid of vowel 'a', then the number of training sessions for vowel 'a' will be increased according to the preset threshold; if the adjustment value of the first formant of the child's vowel 'a' is less than or equal to the first formant of the centroid of vowel 'a', then it will be determined whether the tongue distance is less than the child's target adjustment coefficient for vowel space area × the child's tongue distance parameter. If not, the number of training sessions for vowels o and e is reduced according to the preset threshold, and the process proceeds to step S12; if yes, it is determined whether the adjustment value of the second formant of the child's vowel i is greater than the second formant of the centroid of vowel i. The formula for calculating the adjustment value of the second formant of the vowel i in children is as follows: Adjustment value of the second formant of vowel i in children = Second parameter of vowel i in children + (Target adjustment coefficient of vowel space area in children × Tongue distance parameter in children); If the adjustment value of the second formant of the child's vowel i is greater than the second formant of the centroid of vowel i, then the number of training sessions for vowel i is increased according to the preset threshold; if the adjustment value of the second formant of the child's vowel i is less than or equal to the second formant of the centroid of vowel i, then it is determined whether the adjustment value of the second formant of the child's vowel u is less than the second formant of the centroid of vowel u. The formula for calculating the adjustment value of the second formant of the child vowel u is as follows: Adjustment value of the second formant of the vowel u in children = parameter of vowel u in children - (target adjustment coefficient of vowel space area in children × parameter of tongue distance in children); If the adjustment value of the second formant of the child's vowel u is less than the second formant of the centroid of the vowel u, then the number of training sessions for the vowel u is increased according to a preset threshold, and the number of training sessions for the vowels o and e is decreased according to a preset threshold, proceeding to step S12; if the adjustment value of the second formant of the child's vowel u is greater than or equal to the second formant of the centroid of the vowel u, then the number of training sessions for the vowels o and e is decreased according to a preset threshold, proceeding to step S12. In the second children's vowel practice development mode, the target area of ​​the second vowel space is greater than or equal to the area of ​​the first vowel space. The corresponding training content is: the number of training sessions for vowels i and u is reduced according to a preset threshold, and the number of training sessions for vowels a, o, and e is increased according to a preset threshold, and then proceed to step S12.

8. The method as described in claim 1, characterized in that... In step S10, the formula for calculating the target adjustment coefficient for the vowel space area of ​​males is: Male vowel space area target adjustment factor = second vowel space area target / male adjustment factor; The first model for developing male vowel exercises includes the following steps: Determine whether the mandibular distance is less than the male vowel space area target adjustment coefficient × male mandibular distance parameter; If not, determine whether the tongue distance is less than the male vowel space area target adjustment coefficient × male tongue distance parameter; if yes, determine whether the adjustment value of the first formant of male vowel i is less than the first formant of the centroid of vowel i. The formula for calculating the adjustment value of the first formant of the male vowel i is as follows: Adjustment value of the first formant of male vowel i = First parameter of male vowel i - (Target adjustment coefficient of male vowel space area × Male mandibular distance parameter); If the adjustment value of the first formant of the male vowel i is less than the first formant of the centroid of vowel i, the number of training times for vowel i is increased according to the preset threshold; if the adjustment value of the first formant of the male vowel i is greater than or equal to the first formant of the centroid of vowel i, it is determined whether the adjustment value of the first formant of the male vowel a is greater than the first formant of the centroid of vowel a. The formula for calculating the adjustment value of the first formant of the male vowel 'a' is as follows: The adjustment value of the first formant of male vowel a = male vowel a parameter + (male vowel space area target adjustment coefficient × male mandibular distance parameter); If the adjustment value of the first formant of the male vowel a is greater than the first formant of the centroid of vowel a, then the number of training sessions for vowel a will be increased according to the preset threshold; if the adjustment value of the first formant of the male vowel a is less than or equal to the first formant of the centroid of vowel a, then it is determined whether the tongue distance is less than the male vowel space area target adjustment coefficient × male tongue distance parameter. If not, the number of training sessions for vowels o and e is reduced according to the preset threshold, and the process proceeds to step S12; if yes, it is determined whether the adjustment value of the second formant of the male vowel i is greater than the second formant of the centroid of vowel i. The formula for calculating the adjustment value of the second formant of the male vowel i is as follows: The adjustment value of the second formant of male vowel i = the second parameter of male vowel i + (the target adjustment coefficient of male vowel space area × male tongue distance parameter); If the adjustment value of the second formant of the male vowel i is greater than the second formant of the centroid of vowel i, the number of training times for vowel i is increased according to the preset threshold; if the adjustment value of the second formant of the male vowel i is less than or equal to the second formant of the centroid of vowel i, it is determined whether the adjustment value of the second formant of the male vowel u is less than the second formant of the centroid of vowel u. The formula for calculating the adjustment value of the second formant of the male vowel u is as follows: Adjustment value of the second formant of male vowel u = male vowel u parameter - (male vowel space area target adjustment coefficient × male tongue distance parameter); If the adjustment value of the second formant of the male vowel u is less than the second formant of the centroid of vowel u, then the number of training sessions for vowel u is increased according to a preset threshold, and the number of training sessions for vowels o and e is decreased according to a preset threshold, and proceed to step S12; if the adjustment value of the second formant of the male vowel u is greater than or equal to the second formant of the centroid of vowel u, then the number of training sessions for vowels o and e is decreased according to a preset threshold, and proceed to step S12. In the second male vowel practice design mode, the target area of ​​the second vowel space is greater than or equal to the area of ​​the first vowel space. The corresponding training content is as follows: the number of training sessions for vowels i and u is reduced according to a preset threshold, and the number of training sessions for vowels a, o, and e is increased according to a preset threshold, and then proceed to step S12.

9. The method as described in claim 1, characterized in that... In step S11, the formula for calculating the target adjustment coefficient of the vowel space area for females is: target adjustment coefficient of the vowel space area for females = target area of ​​the second vowel space / female adjustment coefficient; The first model for developing female vowel exercises includes the following steps: Determine whether the mandibular distance is less than the female vowel space area target adjustment coefficient × female mandibular distance parameter; If not, determine whether the tongue distance is less than the female vowel space area target adjustment coefficient × female tongue distance parameter; if yes, determine whether the female vowel i first formant adjustment value is less than the vowel i centroid first formant. The formula for calculating the adjustment value of the first formant of the female vowel i is as follows: The adjustment value of the first formant of the female vowel i = the first parameter of the female vowel i - (the target adjustment coefficient of the female vowel space area × the female mandibular distance parameter); If the adjustment value of the first formant of the female vowel i is less than the first formant of the centroid of vowel i, the number of training times for vowel i is increased according to the preset threshold; if the adjustment value of the first formant of the female vowel i is greater than or equal to the first formant of the centroid of vowel i, it is determined whether the adjustment value of the first formant of the female vowel a is greater than the first formant of the centroid of vowel a. The formula for calculating the adjustment value of the first formant of the female vowel 'a' is as follows: The adjustment value of the first formant of the female vowel a = the parameter of the female vowel a + (the target adjustment coefficient of the female vowel space area × the female mandibular distance parameter). If the adjustment value of the first formant of the female vowel 'a' is greater than the first formant of the centroid of vowel 'a', then the number of training sessions for vowel 'a' will be increased according to the preset threshold; if the adjustment value of the first formant of the female vowel 'a' is less than or equal to the first formant of the centroid of vowel 'a', then it will be determined whether the tongue distance is less than the female vowel space area target adjustment coefficient × female tongue distance parameter. If not, the number of training sessions for vowels o and e is reduced according to the preset threshold, and the process proceeds to step S12; if yes, it is determined whether the adjustment value of the second formant of the female vowel i is greater than the second formant of the centroid of vowel i. The formula for calculating the adjustment value of the second formant of the female vowel i is as follows: The adjustment value of the second formant of the female vowel i = the second parameter of the female vowel i + (the target adjustment coefficient of the female vowel space area × the female tongue distance parameter); If the adjustment value of the second formant of the female vowel i is greater than the second formant of the centroid of vowel i, the number of training times for vowel i is increased according to the preset threshold; if the adjustment value of the second formant of the female vowel i is less than or equal to the second formant of the centroid of vowel i, it is determined whether the adjustment value of the second formant of the female vowel u is less than the second formant of the centroid of vowel u. The formula for calculating the adjustment value of the second formant of the female vowel u is as follows: The adjustment value of the second formant of the female vowel u = the female vowel u parameter - (the female vowel space area target adjustment coefficient × the female tongue distance parameter); If the adjustment value of the second formant of the female vowel u is less than the second formant of the centroid of vowel u, then the number of training sessions for vowel u is increased according to a preset threshold, and the number of training sessions for vowels o and e is decreased according to a preset threshold, and proceed to step S12; if the adjustment value of the second formant of the female vowel u is greater than or equal to the second formant of the centroid of vowel u, then the number of training sessions for vowels o and e is decreased according to a preset threshold, and proceed to step S12. In the second female vowel practice design mode, the target area of ​​the second vowel space is greater than or equal to the area of ​​the first vowel space. The corresponding training content is as follows: the number of training sessions for vowels i and u is reduced according to a preset threshold, and the number of training sessions for vowels a, o, and e is increased according to a preset threshold, and then proceed to step S12.

10. A system for multi-dimensional adaptive speech training, characterized in that... It includes a processor and a memory; wherein the memory is coupled to the processor and is used to store a computer program that, when executed by the processor, causes the processor to implement the method described in any one of claims 1 to 9.

Images