[go: up one dir, main page]

CN119418699B - An intelligent voice interaction system and electronic student ID card - Google Patents

An intelligent voice interaction system and electronic student ID card Download PDF

Info

Publication number
CN119418699B
CN119418699B CN202510006424.7A CN202510006424A CN119418699B CN 119418699 B CN119418699 B CN 119418699B CN 202510006424 A CN202510006424 A CN 202510006424A CN 119418699 B CN119418699 B CN 119418699B
Authority
CN
China
Prior art keywords
task
data
analysis
tasks
semantic
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202510006424.7A
Other languages
Chinese (zh)
Other versions
CN119418699A (en
Inventor
温仁宝
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Fujian Fuyu Wisdom Information Technology Co ltd
Original Assignee
Fujian Fuyu Wisdom Information Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Fujian Fuyu Wisdom Information Technology Co ltd filed Critical Fujian Fuyu Wisdom Information Technology Co ltd
Priority to CN202510006424.7A priority Critical patent/CN119418699B/en
Publication of CN119418699A publication Critical patent/CN119418699A/en
Application granted granted Critical
Publication of CN119418699B publication Critical patent/CN119418699B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/205—Parsing
    • G06F40/211—Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/253—Grammatical analysis; Style critique
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/279—Recognition of textual entities
    • G06F40/284—Lexical analysis, e.g. tokenisation or collocates
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/30—Semantic analysis
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/48—Program initiating; Program switching, e.g. by interrupt
    • G06F9/4806—Task transfer initiation or dispatching
    • G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
    • G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L13/00—Speech synthesis; Text to speech systems
    • G10L13/08—Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/26—Speech to text systems
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
    • G10L2015/225—Feedback of the input speech

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computational Linguistics (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Health & Medical Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • General Health & Medical Sciences (AREA)
  • Multimedia (AREA)
  • Acoustics & Sound (AREA)
  • Human Computer Interaction (AREA)
  • Data Mining & Analysis (AREA)
  • Software Systems (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Machine Translation (AREA)

Abstract

本发明涉及智能语音交互技术领域，具体为一种智能语音交互系统及电子学生证，系统包括语音输入处理模块、语义分解模块、任务分配模块、逻辑执行模块、动态响应模块、语音交互调整模块。本发明中，通过解析语音信号的频域与时域特征，精准捕获语音细节，提升识别准确性，逐词解析与重组语义内容，支持复杂语义的准确提取与任务描述，资源消耗与依赖关系分析优化任务顺序，提升系统高效性与适应性，任务执行结合数据检索与内容分类，确保关键任务与关联任务协调进行，动态反馈标记未完成任务并组装补充信息，增强对话连贯性与响应自然性，语音交互调整模块分析未完成指令与重复调用，提升对需求变化的适配性和服务效率。

The present invention relates to the field of intelligent voice interaction technology, specifically to an intelligent voice interaction system and an electronic student ID card, wherein the system includes a voice input processing module, a semantic decomposition module, a task allocation module, a logic execution module, a dynamic response module, and a voice interaction adjustment module. In the present invention, by analyzing the frequency domain and time domain features of the voice signal, the voice details are accurately captured, the recognition accuracy is improved, the semantic content is parsed and reorganized word by word, the accurate extraction and task description of complex semantics are supported, the resource consumption and dependency relationship analysis optimize the task sequence, the system efficiency and adaptability are improved, the task execution is combined with data retrieval and content classification, and the key tasks and related tasks are coordinated. Dynamic feedback marks unfinished tasks and assembles supplementary information to enhance the dialogue coherence and natural response. The voice interaction adjustment module analyzes unfinished instructions and repeated calls to improve the adaptability to demand changes and service efficiency.

Description

Intelligent voice interaction system and electronic student identity card
Technical Field
The invention relates to the technical field of intelligent voice interaction, in particular to an intelligent voice interaction system and an electronic student identity card.
Background
The field of intelligent speech interaction technology involves the use of speech recognition, natural Language Processing (NLP), dialog management, and speech synthesis techniques to enable interactions between humans and computer systems, the core of which is to allow users to communicate in natural language, while the system is able to parse intent and respond, including extracting key information from user speech input, processing context, and persistence of dialog, and generating natural and accurate speech responses. The intelligent voice interaction technology plays a key role in improving user experience and enhancing accessibility of a human-computer interface, and is widely applied to multiple fields of intelligent assistants, vehicle-mounted systems, intelligent home control, customer service and the like.
The system can identify and respond to one-time instructions, can maintain topic continuity according to conversation history, process complex interaction and adapt to user demand change, is included in an application scene of an electronic student identity card, can verify identity, provide course information, manage courses, test arrangement and the like through voice interaction, and utilizes the direction of the electronic student identity card, so that students can access personal academics through simple voice commands, and the service efficiency of educational institutions and the convenience of the students are greatly improved.
In the prior art, the detail characteristic of a voice signal is not captured sufficiently in the voice recognition process, so that the content recognition in nonstandard voice or background noise is inaccurate, in semantic analysis, a semantic unit and a limiting condition are difficult to separate accurately, the understanding of a complex command is not comprehensive enough, the logic dependency relationship and time limit analysis of a task are insufficient, in the aspect of task distribution, the accurate evaluation of task priority and the careful analysis of dependency sequence are lacking, the execution sequence disorder or the resource allocation is easy to be unreasonable, in the dynamic feedback process, the integration capability of the prior art on context information is limited, the problem that feedback content is separated from the intention of a user occurs, the continuity of multi-round interaction is influenced, the repeated call and incomplete task processing is not intelligent enough in multi-round dialogue, the interaction efficiency is low, the user experience is influenced, the problems of response delay, task execution error, user satisfaction degree reduction and the like are caused in the practical application, and the wide adaptability and the accurate service capability of an intelligent voice interaction system are restricted.
Disclosure of Invention
The invention aims to solve the defects in the prior art, and provides an intelligent voice interaction system and an electronic student identity card.
In order to achieve the above purpose, the invention adopts the following technical scheme that an intelligent voice interaction system comprises:
the voice input processing module divides an audio frame by analyzing the frequency domain characteristics and the time domain characteristics of a voice signal, extracts amplitude and frequency distribution, recognizes a morpheme sequence, integrates a time relationship and generates a voice text sequence;
the semantic decomposition module analyzes word-by-word part of the speech text sequence, separates verbs and target objects, extracts time and range conditions, associates semantic logic and generates a task semantic description set;
The task distribution module analyzes the key tasks and the secondary tasks in the task semantic description set, analyzes the resource and the dependency sequence, prioritizes and sorts the groups and distributes the sequence, decomposes the chains and generates a task scheduling list;
The logic execution module loads task attributes in the task scheduling list, calls an electronic student identity retrieval course, analyzes the dependence, classifies and integrates the dependence, executes inquiry tasks and generates a task execution result set;
the dynamic response module analyzes the completed query content based on the task execution result set, marks the incomplete task, supplements the context as a feedback text, converts the voice signal output and acquires user feedback audio;
And the voice interaction adjusting module extracts dialogue content and task execution records based on the user feedback audio, analyzes incomplete instructions, adjusts a task chain and distributes a flow, and generates a voice interaction task distribution data table.
As a further scheme of the invention, the voice text sequence comprises an amplitude characteristic, a frequency distribution characteristic, a morpheme sequence and a time relation, the task semantic description set comprises a time limiting condition, a range limiting condition, a logic relation and a key semantic object, the task scheduling list comprises a key task, a secondary task, a task priority, a task dependency sequence and a resource consumption evaluation, the task execution result set comprises a query operation result, task classification data and an association condition integration result, the feedback text comprises context information, natural language feedback content and an incomplete task mark, the user feedback audio comprises a query content, an incomplete instruction and a new instruction content, and the voice interaction task allocation data table comprises an incomplete instruction, a repeated call record, task chain adjustment data and call flow allocation information.
As a further aspect of the present invention, the voice input processing module includes:
The signal analysis sub-module analyzes the frequency domain characteristics and the time domain characteristics of the voice signal, performs quantitative analysis on the energy distribution of the signal, performs periodic detection and amplitude analysis on the waveform, and adjusts the detail parameters of the frequency domain analysis to obtain signal characteristic description;
The frame segmentation sub-module adopts the signal characteristic description to carry out frame segmentation of the audio signal, adjusts the time window and interval parameters of the segmentation, optimizes the continuity and resolution of the data in the frame, extracts the amplitude, frequency and energy of each frame, and acquires a frame characteristic data set through the analysis of the frame characteristic data;
The morpheme integration submodule utilizes the frame characteristic data set to carry out morpheme analysis and integration of continuous frames, strengthens logic and structural relation among morphemes through time sequence analysis, optimizes consistency of morphemes and generates a voice text sequence.
As a further aspect of the present invention, the semantic decomposition module includes:
The part-of-speech analysis submodule carries out part-of-speech tagging based on the voice text sequence, comprises the identification of verbs and nouns, uses database data to carry out comparison analysis to verify the part of speech of each word, adjusts analysis precision and matches text complexity, and acquires part-of-speech relation data;
The syntactic reorganization sub-module adopts the part-of-speech relation data, adjusts the relation between verbs and target objects according to syntactic rules, optimizes sentence structure, analyzes key syntactic elements and reorganizes sentence components to obtain syntactic structure data;
based on the syntactic structure data, the logic relation submodule adopts a conditional semantic nesting analysis method to extract time and range limiting conditions in the sentence, associates the conditions with the sentence key semantic objects, integrates the key semantic elements by using the logic analysis method, and generates a task semantic description set.
As a further scheme of the invention, the formula adopting the conditional semantic nesting analysis method is as follows:
;
wherein, Represents the set of values of the task semantic descriptions,A normalized value representing a time constraint in a sentence,Representing the semantic density of the scope defining conditions in the sentence,Representing the associative matching degree of key semantic objects in the sentence,The sum of weights representing key semantic elements in the logic analysis process,Correction coefficients representing the hierarchy in the syntactic structure data,A distribution complexity index representing conditions and key semantic objects in the statement,、In order to adjust the coefficient of the power supply,、Is a normalized coefficient.
As a further aspect of the present invention, the task allocation module includes:
The task analysis submodule identifies and classifies the key and non-key tasks based on the task semantic description set, evaluates the resource and time consumption of each task, optimizes the precision of resource allocation and time allocation, adjusts the urgency of the evaluation parameter matching differentiation task, and acquires task resource consumption data;
The dependency sequencing sub-module uses the task resource consumption data to analyze the dependency relationship and the execution priority among tasks, optimizes the task execution sequence by adjusting the priority sequencing parameters, and generates a task dependency sequence;
And the task scheduling submodule formulates task execution groups and sequences according to the task dependency sequences, plans the scheduling of each task, adjusts the execution grouping parameters and scheduling strategies and matches the requirements of the tasks, and generates a task scheduling list.
As a further aspect of the present invention, the logic execution module includes:
The task attribute loading submodule identifies the attribute of each task based on the task scheduling list, analyzes and loads the task attribute, refines the loading flow by adopting a real-time adjustment technology, adjusts the loading rate and optimizes task matching logic, and generates a task attribute data set;
the course information retrieval submodule uses the task attribute data set to call the electronic student identity data to retrieve target course information, and comprises the steps of adjusting and optimizing retrieval parameters, extracting key course dependent information and obtaining course dependent data;
And the dependence analysis sub-module integrates the key tasks and the associated conditional tasks through the course dependence data, adjusts the query parameters, and organizes and executes the query parameters in sequence to generate a task execution result set.
As a further aspect of the present invention, the dynamic response module includes:
the query result analysis sub-module analyzes the query result based on the task execution result set, distinguishes completed and unfinished query tasks by adjusting analysis parameters, optimizes data screening logic, and acquires voice query analysis data;
the new instruction marking sub-module uses the voice query analysis data to mark new instructions for the incomplete query task, optimizes marking flow, strengthens the response speed of recognition, adjusts classification parameters and generates a new instruction task set;
And the natural language feedback submodule gathers natural language feedback aiming at the user according to the new instruction task set, adjusts feedback parameters, converts feedback data into voice signals, optimizes a voice conversion process, captures feedback audio of the user and obtains user feedback audio.
As a further aspect of the present invention, the voice interaction adjustment module includes:
The dialogue analysis submodule analyzes dialogue content including data related to the electronic student card task execution record based on the user feedback audio, distinguishes and identifies incomplete and repeated instructions by optimizing the conversion precision of voice to text, strengthens analysis of detail processing and generates a detail dialogue analysis result;
The task chain reorganization sub-module uses the detail dialogue analysis result to reorder the tasks identified in the dialogue, adjust the execution sequence of the tasks and reflect the priority of real-time user demands, and fine-tune task management logic to optimize the execution path so as to acquire reorganized task chain data;
and the flow reassignment sub-module reassigns the invoking flow of the electronic student identity function according to the reorganization task chain data, adjusts the invoking sequence and resource assignment of the tasks, matches the user interaction requirement, optimizes the invoking flow and generates a voice interaction task assignment data table.
Based on the same inventive concept, an electronic student identity card is also provided, which comprises a memory and a processor, wherein the memory stores a computer program, and the processor realizes the intelligent voice interaction system when executing the computer program.
Compared with the prior art, the invention has the advantages and positive effects that:
According to the invention, the audio signal is converted into the voice text sequence by analyzing the frequency domain and time domain characteristics of the voice signal, thereby being beneficial to accurately capturing the detail characteristics in the voice, improving the recognition accuracy, on the aspect of processing semantic content, establishing the logical relationship between time, range and semantic objects by analyzing and recombining the voice text sequence word by word, ensuring the clear information structure, supporting the accurate extraction of complex semantics and task description, optimizing the task execution sequence by analyzing the resource consumption and the dependency relationship in the task distribution link, enhancing the high efficiency and the adaptability of the system, integrating the task execution process by combining data retrieval and the dependency content classification, and ensuring the execution coordination of key tasks and associated tasks. The dynamic feedback mechanism improves the consistency of conversation and the naturalness of response by assembling marks and supplementary information of incomplete tasks, the voice interaction adjustment module analyzes incomplete instructions and repeated calling content, optimizes task allocation and flow arrangement, enhances the sensitivity and coping capacity to user demand change, and enhances the voice input and processing, task decomposition and scheduling, dynamic adjustment and feedback generation capacity, so that the intelligent voice interaction system has higher accuracy, consistency and suitability, and simultaneously remarkably improves the user experience and the service efficiency of the system.
Drawings
FIG. 1 is a flow chart of the system of the present invention.
FIG. 2 is a schematic diagram of a system framework of the present invention.
FIG. 3 is a flow chart of a speech input processing module according to the present invention.
FIG. 4 is a flow chart of a semantic decomposition module according to the present invention.
FIG. 5 is a flow chart of a task allocation module according to the present invention.
FIG. 6 is a flow chart of a logic execution module according to the present invention.
FIG. 7 is a flow chart of the dynamic response module of the present invention.
FIG. 8 is a flow chart of a voice interaction adjusting module according to the present invention.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Referring to fig. 1 to 2, an intelligent voice interaction system includes:
the voice input processing module performs frame segmentation on the audio signal by analyzing the frequency domain characteristics and the time domain characteristics of the voice signal, extracts the amplitude and frequency distribution characteristics in continuous frames, identifies the morpheme sequence, integrates the time relationship and the structural content among morphemes, and generates a voice text sequence;
The semantic decomposition module analyzes word-by-word relation of the voice text sequence, separates verbs and target objects, reorganizes semantics through syntactic rules, extracts time and range limiting conditions, associates logic relation between limiting conditions and key semantic objects, combines analysis contents, and generates task semantic description sets;
the task allocation module analyzes the key tasks and the secondary tasks in the task semantic description set, analyzes the resource consumption time of each task, evaluates the dependency sequence among the tasks, allocates the groups and the sequences of task execution according to the priority order, and decomposes the tasks according to the dependency chain order to generate a task scheduling list;
the logic execution module loads task attributes in the task scheduling list, invokes the electronic student identity data to retrieve course information, analyzes task dependent content in the data, classifies and integrates the task dependent content, executes inquiry operation required by a key task, integrates associated condition tasks in sequence, and generates a task execution result set;
The dynamic response module analyzes the completed query content based on the task execution result set, marks the incomplete task in the user request as a new instruction, assembles context information through a content supplementing mechanism, assembles a feedback text in a natural language form, converts the text into a voice signal and outputs the voice signal, and acquires user feedback audio;
The voice interaction adjusting module extracts dialogue content and task execution records of the electronic student certificates in the feedback audio of the user, analyzes unfinished instructions in the dialogue and repeatedly called electronic student certificate functions, rearranges task chains, and distributes calling flows to generate a voice interaction task distribution data table.
The voice text sequence comprises amplitude characteristics, frequency distribution characteristics, morpheme sequences and time relations, the task semantic description set comprises time limiting conditions, range limiting conditions, logic relations and key semantic objects, the task scheduling list comprises key tasks, secondary tasks, task priorities, task dependence sequences and resource consumption evaluation, the task execution result set comprises query operation results, task classification data and association condition integration results, the feedback text comprises context information, natural language feedback content and incomplete task marks, the user feedback audio comprises query content, incomplete instructions and new instruction content, the voice interaction task allocation data table comprises incomplete instructions, repeated call records, task chain adjustment data and call flow allocation information.
Specifically, as shown in fig. 2 and 3, the voice input processing module includes:
The signal analysis submodule analyzes the frequency domain characteristics and the time domain characteristics of the voice signal, carries out quantitative analysis on the energy distribution of the signal, carries out periodic detection and amplitude analysis on the waveform, and adjusts the detail parameters of the frequency domain analysis to obtain the execution flow of signal characteristic description as follows;
Analyzing the frequency domain characteristics and the time domain characteristics of the voice signal, carrying out quantitative analysis on the energy distribution of the signal through the analysis characteristics, wherein the analysis relates to the density calculation of a frequency spectrum and the total energy evaluation, the detail parameter adjustment in the frequency domain analysis is used for reflecting the real characteristics of the signal more accurately, the adjusted parameters are optimized through a mathematical optimization method such as gradient descent, the obtained frequency domain representation is ensured to be closest to the original signal, the optimization process needs to update the parameters according to the feedback iteration of the actual signal, the periodic detection comprises the measurement of the signal period and the amplitude analysis, and the periodic change in the signal is used for identifying the periodic change, which is particularly important for a voice recognition system, because the voice mode and the voice speed of a speaker can be revealed, and the signal characteristic description comprehensively refers to all analysis results of the signal, so that necessary data support is provided for subsequent processing.
The frame segmentation sub-module adopts signal characteristic description to carry out frame segmentation of the audio signal, adjusts the time window and interval parameters of the segmentation, optimizes the continuity and resolution of data in frames, extracts the amplitude, frequency and energy of each frame, and obtains the execution flow of a frame characteristic data set through the analysis of the frame characteristic data as follows;
Frame segmentation of an audio signal is performed using signal characteristic descriptions, the time window and interval of the frames are adjusted according to the signal characteristics, and the following formula is used:
;
The duration of the frame is calculated, wherein, The time window of the frame is indicated,Representing the number of samples in a frame,Is the sampling frequency;
the formula assists in precisely controlling the time length and coverage of the frame, so as to optimize the continuity and resolution of data in the frame, and a frame characteristic data set is calculated by amplitude, frequency and energy data in the frame;
By actual monitoring, the sampling rate of the signal is assumed 44.1KHz, number of samples per frameSet to 1024, and substituted into the formula to obtainSecond, this indicates that each frame lasts about 23.2 milliseconds, and this arrangement effectively captures short-time variations in the speech signal in practical applications, facilitating subsequent speech analysis and feature extraction.
The morpheme integration submodule utilizes a frame characteristic data set to carry out morpheme analysis and integration of continuous frames, strengthens logic and structural relation among morphemes through time sequence analysis, optimizes consistency of morphemes, and generates an execution flow of a voice text sequence as follows;
The method comprises the steps of carrying out morpheme analysis and integration of continuous frames by using a frame characteristic data set, wherein the integration basis of morphemes is characteristic data comparison among the continuous frames, processing data by a time sequence analysis method, such as an autoregressive moving average model or a hidden Markov model, strengthening logic and structural relations among morphemes, improving the consistency of morphemes by the analysis, enabling the integrated morphemes to be more consistent with the semantic and grammar structures of natural language, enabling the generated voice text sequence to be smoother and more natural, optimizing the whole semantic expression, and ensuring that the voice recognition system can accurately convert voice input into text output by the processing of each morpheme closely depending on the analysis result of the previous step in the process of real-time feedback adjustment.
Specifically, as shown in fig. 2 and 4, the semantic decomposition module includes:
The part-of-speech analysis submodule carries out part-of-speech tagging based on a voice text sequence, comprises the identification of verbs and nouns, carries out comparison analysis to verify the part of speech of each word by using database data, adjusts analysis precision and matches text complexity, and an execution flow for acquiring part-of-speech relation data is as follows;
The recognition of verbs and nouns in the process depends on corpus data, the data provides necessary references to verify the part of speech of each word, a sub-module can determine the proper part of speech of words in different contexts by comparing typical use cases in a word stock, the adjustment of analysis accuracy is realized by algorithm iteration, the recognition and correction of wrong part of speech markers are included, the complexity of matching texts involves the analysis of the structural and grammatical diversity of texts, the system is ensured to adapt to simple to complex text input, the acquired part of speech relation data is the basis of subsequent syntactic reorganization, and the data reflects the grammatical relation and structural layout between words.
The syntactic reorganization submodule adopts part-of-speech relation data, adjusts the relation between verbs and target objects according to syntactic rules, optimizes sentence structure, analyzes key syntactic elements and reorganizes sentence components, and obtains the execution flow of the syntactic structure data as follows;
The method comprises the steps of adjusting the relation between verbs and target objects according to a syntactic rule by using word relation data, analyzing key syntactic elements and reorganizing sentence components in a syntactic reorganization process, wherein the method relates to a complex sentence component analysis and reorganization technology, can optimize the structure of sentences, improves the fluency and logical continuity of sentences, and is characterized in that the syntactic structure data is a basis for generating more accurate semantic descriptions, the data integrates all syntactic analysis results and provides detailed syntactic relation and structure information, which is a key for understanding and generating complex texts.
The logic relation sub-module extracts time and range limiting conditions in the sentence based on the syntactic structure data by adopting a conditional semantic nesting analysis method, associates the conditions with sentence key semantic objects, integrates the key semantic elements by using the logic analysis method, and generates an execution flow of a task semantic description set as follows;
The formula for the conditional semantic nesting analysis method is as follows:
;
wherein, Represents the set of values of the task semantic descriptions,A normalized value representing a time constraint in a sentence,Representing the semantic density of the scope defining conditions in the sentence,Representing the associative matching degree of key semantic objects in the sentence,The sum of weights representing key semantic elements in the logic analysis process,Correction coefficients representing the hierarchy in the syntactic structure data,A distribution complexity index representing conditions and key semantic objects in the statement,、In order to adjust the coefficient of the power supply,、Is a normalized coefficient;
Formula details and formula calculation derivation process:
the normalized value representing the time constraints extracted from the sentence, in practical application, assuming 100 text paragraphs are analyzed, the time conditions vary from date to specific hour, and the normalized values are normalized, such as by using the Z-score method, Is 1.5, representing a deviation from the mean time condition of 1.5 standard deviation;
Semantic density representing the range limiting conditions extracted from the sentence, if the sentence contains 10 range conditions such as 'national', 'regional', 'specific city' and the like, through frequency analysis and inverse document frequency calculation, A value of 0.8;
The association matching degree refers to the fact that 80% of matches are mined through comparison with known semantic objects in a database, and therefore the matching degree is set to be 0.8;
The total weight of key semantic elements in the logic analysis process is 2.5 if 5 key elements exist, the weight of each key element is set according to the previous research result;
the correction coefficients of the hierarchy in the syntactic structure data are adjusted according to sentence complexity, provided that the sentence is more complex, Set to 0.7;
The distribution complexity index of the condition and the key semantic object is determined to be 1.2 after calculation;
And Adjusting the coefficients to be 0.5 and 0.3 respectively, and obtaining an optimized value according to the previous experiments and optimization for multiple times;
And The normalization coefficient is set to be 1 and 0.5 so as to ensure that the calculation result of the formula is in a reasonable range;
calculating the square root of the absolute value of the product of the time condition and the range condition: ;
Calculating the contribution of the key semantic objects after the association matching degree adjustment: ;
Calculating the absolute value adjusted contribution of the weight difference in the logic analysis: ;
Calculating normalized and complexity adjusted values of denominators: ;
to sum up, the final calculation of the formula is:
;
the result shows that by means of a conditional semantic nesting analysis method, key semantic elements are effectively integrated according to semantic density of time and range and matching degree of the semantic density and a logic relation, and the obtained task semantic description set value is about 1.273, so that the calculation method can effectively integrate input sentence conditions, a relatively accurate semantic description set is generated, and application effects and accuracy of an analysis model are reflected.
Specifically, as shown in fig. 2 and 5, the task allocation module includes:
the task analysis submodule identifies and classifies critical and non-critical tasks based on a task semantic description set, evaluates the resource and time consumption of each task, optimizes the precision of resource allocation and time allocation, adjusts the urgency of matching evaluation parameters with differentiated tasks, and acquires the execution flow of task resource consumption data as follows;
Optimizing the task through single resource consumption coefficient and corresponding time evaluation, and using the formula:
;
the total resource consumption is calculated, wherein, Indicating the total resource consumption of the device,Represent the firstThe resource consumption coefficient of the task in question,Representing a corresponding time;
the total resource amount required for completing a series of tasks can be accurately estimated through the product accumulation of the resource consumption coefficient and the time, so that the resource allocation and the time arrangement are optimized;
referring to a practical example, there are three tasks whose resource consumption coefficients are respectively Corresponding time consumption isThe time period of the time period,The time period of the time period,Hours;
Total resource consumption The calculation is as follows: The result shows that the total of 12 units of resources are needed for completing the three tasks, which is helpful for a manager to reasonably arrange the resources and ensure that the tasks are completed on time without wasting the resources.
The dependency sequencing sub-module uses task resource consumption data to analyze the dependency relationship and execution priority among tasks, optimizes the task execution sequence by adjusting the priority sequencing parameters, and generates the execution flow of a task dependency sequence as follows;
The method comprises the steps of analyzing the dependency relationship and the execution priority among tasks, wherein the process involves identifying the sequence and the resource flow direction among the tasks, optimizing the logic and the efficiency of task execution through data analysis and priority adjustment, adopting a sequencing algorithm such as a topology sequencing or priority queue technology to ensure that preconditions are met, activating the subsequent tasks, optimizing the execution sequence, remarkably improving the overall workflow efficiency, reducing waiting and idle time, adjusting priority sequencing parameters based on the urgency and the complexity of the tasks, and realizing such parameter adjustment through algorithm iteration and feedback adjustment to ensure the smoothness of task execution and on-demand resource allocation.
The task scheduling submodule formulates task execution groups and sequences according to task dependency sequences, plans scheduling of each task, adjusts execution grouping parameters and scheduling strategies and matches requirements of the tasks, and generates an execution flow of a task scheduling list as follows;
the scheduling of each task is planned in detail according to the task dependency sequence, the scheduling policy is set according to the urgency, importance and resource availability of the task, the sub-module adjusts the parameters of the scheduling grouping to match various task requirements, such as parallel processing capacity and interface requirements among the tasks, the generated task scheduling list provides an execution guide for management, the starting time, the execution sequence and the resource allocation of each task are listed in detail, the process ensures that all the tasks can be executed in an optimal time window, thus the overall working efficiency is improved, the resource waste caused by improper scheduling is reduced, the scheduling sub-module is particularly critical to the complex multi-task environment, and the scheduling sub-module utilizes advanced algorithms to ensure the optimization of the scheduling policy, such as genetic algorithm or simulated annealing algorithm for the optimization of the scheduling policy, so as to realize the most effective resource sharing and time coordination among the tasks.
Specifically, as shown in fig. 2 and 6, the logic execution module includes:
The task attribute loading submodule identifies the attribute of each task based on a task scheduling list, analyzes and loads the task attribute, refines the loading flow by adopting a real-time adjustment technology, adjusts the loading rate and optimizes task matching logic, and generates an execution flow of a task attribute data set as follows;
the loading flow is adjusted through a real-time technology, the task matching logic is optimized, and the following formula is used:
;
to adjust the loading rate, wherein, Which is indicative of the loading power of the load,Representing the amount of data to be loaded,Time is;
the formula is used for calculating and adjusting the data loading rate, so that the task attribute data set is ensured to be loaded rapidly and accurately;
referring to an actual scene, the task data volume 500MB, predetermined loading timeFor 2 minutes, the loading power is calculated according to a formulaMB/min, and loading data at a rate of 250MB per minute in a given time, can meet the requirement of the system for rapid data processing, and simultaneously maintain the efficiency and response speed of the system.
The course information retrieval submodule uses the task attribute data set to call the electronic student identity data to retrieve the target course information, and comprises the steps of adjusting and optimizing retrieval parameters, extracting key course dependent information and obtaining the execution flow of the course dependent data as follows;
the electronic student certificate data is called to carry out course information retrieval by utilizing the task attribute data set, the retrieval parameters are adjusted and optimized based on the result of data analysis, the optimization purpose is to improve the retrieval accuracy and efficiency, the adjusted retrieval parameters can reflect the student demands and course dependency relations more accurately, the extraction of the key course dependency information focuses on analyzing the logic relations among courses, the method has important significance for student course planning and teaching resource allocation, the information of the courses required by students can be screened efficiently by refining the retrieval logic and the optimization algorithm, and data support is provided for teaching management and student learning.
The dependence analysis sub-module integrates the key tasks and the associated conditional tasks through course dependence data, adjusts the query parameters, and organizes and executes the query parameters in sequence to generate an execution flow of a task execution result set as follows;
Integrating the key tasks and the associated conditional tasks, adjusting the query parameters to optimize the task execution flow, analyzing the dependency relationship between the tasks by an algorithm and setting the execution sequence, wherein the key is to ensure that all the tasks are correctly executed according to logic and priority, dynamically adjusting the query parameters, the module can flexibly respond to the changing task requirements and condition limitations, the organization of the execution sequence involves complex logic judgment and parameter setting, the coordination among tasks and the consistency of data are ensured, and the generated task execution result set provides execution benchmarks and references for subsequent tasks and supports the stable operation and efficient management of the system.
Specifically, as shown in fig. 2 and 7, the dynamic response module includes:
The query result analysis submodule analyzes the query result based on the task execution result set, distinguishes completed and unfinished query tasks by adjusting analysis parameters, optimizes data screening logic, and obtains the execution flow of voice query analysis data as follows;
By adjusting the analysis parameters, the completed and incomplete query tasks are distinguished, using the formula:
;
wherein, The percentage of completion of the task is indicated,Is the number of tasks that are not completed,Is the total number of tasks;
The calculation model assists in evaluating the overall task progress, so that the effectiveness and optimization of the data screening logic are ensured;
Setting the total task number 100, Number of incomplete tasks20, The task completion degreeCalculated asCurrently 80% of the tasks are completed, an index that is critical to assessing project progress and adjusting future strategies, and helps the manager identify areas that require additional resources and attention.
The new instruction marking sub-module analyzes data by using voice query, carries out new instruction marking on the incomplete query task, optimizes marking flow, strengthens the response speed of recognition, adjusts classification parameters and generates an execution flow of a new instruction task set as follows;
The voice query analysis data are utilized to carry out marking and classification of new instructions, the adjusted classification parameters are helpful for accurately distinguishing the states of various query tasks, incomplete query tasks receive new instruction marks through an optimized flow, in the marking process, the system analyzes and identifies task keywords through an algorithm, marking logic is adjusted to improve marking accuracy and response speed, adjustment in the system comprises learning and adapting to query habits of users, and frequent query problems are processed preferentially, the optimized result not only improves the operation efficiency of the system, but also enhances user experience, and all incomplete tasks are ensured to be processed timely according to priority and resource availability.
The natural language feedback sub-module collects natural language feedback aiming at a user according to a new instruction task set, adjusts feedback parameters, converts feedback data into voice signals, optimizes a voice conversion process, captures feedback audio of the user, and obtains an execution flow of the feedback audio of the user as follows;
according to the new instruction task set, the natural language feedback of the user is summarized, the feedback parameters are adjusted to enhance the accuracy and fluency of voice recognition, the text feedback is converted into voice signals through advanced voice recognition technology and natural language processing algorithm, the optimization process comprises adjustment of tone, speed and rhythm of voice so as to adapt to pronunciation characteristics of different users, the noise filtering and voice enhancement functions are added by capturing user feedback audio, so that the instructions of the user are accurately captured and analyzed in a noisy environment, the improved feedback mechanism enables interaction between the user and the system to be more natural and efficient, and the collected audio data is also used for further training and optimizing a voice recognition model of the system.
Specifically, as shown in fig. 2 and 8, the voice interaction adjusting module includes:
The dialogue analysis submodule analyzes dialogue content based on user feedback audio, comprises data related to the task execution record of the electronic student card, distinguishes and identifies incomplete and repeated instructions by optimizing the conversion precision of voice to text, strengthens analysis on detail processing, and generates an execution flow of detail dialogue analysis results as follows;
Based on user feedback audio analysis dialogue content, voice data is converted into text data, then the text is analyzed to identify incomplete and repeated instructions, the technology used in the conversion process is an advanced voice recognition algorithm, the algorithm optimizes recognition accuracy through a large amount of data training, the voice-to-text conversion accuracy is optimized, the accuracy of voice recognition is improved, recognition of dialects and accents is also involved, and speaker separation technology in a multi-person dialogue scene can accurately distinguish various instructions, repeated instructions are marked to reduce redundant operation of the system, analysis of detail processing further comprises emotion analysis of dialogue content to judge emotion and intention of a user, and guidance is provided for system upgrading in a future time period.
The task chain reorganization sub-module uses the detail dialogue analysis result to reorder the tasks identified in the dialogue, adjusts the execution sequence of the tasks and reflects the priority of real-time user demands, and fine-tunes task management logic to optimize the execution path, and the execution flow of the reorganized task chain data is obtained as follows;
The main method for optimizing the execution path of the fine tuning task management logic comprises the steps of utilizing a decision tree and a priority queue algorithm, automatically adjusting the task sequence according to the priority and the dependency of the tasks by the algorithm, so that the execution path is optimized, the changing requirement of a user can be responded quickly, the generated recombined task chain data enables the system to be more flexible and respond to the operation of the user, and the task management logic is matched closely with the actual interaction of the user.
The flow reassignment submodule reassigns the calling flow of the electronic student identity function according to the reorganized task chain data, adjusts the calling sequence and resource assignment of the tasks, matches the user interaction requirement, optimizes the calling flow and generates the execution flow of the voice interaction task assignment data table as follows;
and (3) carrying out call flow reassignment of the electronic student identity function according to the reorganization task chain data, and using the formula:
;
To optimize the task call flow, wherein, The processing time of the task is indicated,Representing the total amount of resources available,Representing the number of tasks to be processed simultaneously;
The formula reduces the total time of task processing through the optimization of resource allocation, and improves the processing efficiency;
setting a resource amount 1000 Units, number of tasks10, Task processing timeThe calculation is as follows: A unit time;
the result shows that each task obtains 100 units of resources on average, and the system is allowed to process a plurality of tasks in parallel with high efficiency, so that the interaction requirement of a user is responded quickly, and the optimized call flow ensures reasonable allocation of resources and high efficiency of task processing.
An electronic student identity card comprises a memory and a processor, wherein a computer program is stored in the memory, and the processor realizes the intelligent voice interaction system when executing the computer program.
The present invention is not limited to the above embodiments, and any equivalent embodiments which can be changed or modified by the technical disclosure described above can be applied to other fields, but any simple modification, equivalent changes and modification made to the above embodiments according to the technical matter of the present invention will still fall within the scope of the technical disclosure.

Claims (10)

1.一种智能语音交互系统，其特征在于，所述系统包括：1. An intelligent voice interaction system, characterized in that the system comprises: 语音输入处理模块通过解析语音信号的频域特征和时域特征，分割音频帧，提取振幅和频率分布，识别语素序列，整合时间关系，生成语音文本序列；The speech input processing module analyzes the frequency domain and time domain features of the speech signal, segments the audio frames, extracts the amplitude and frequency distribution, identifies the morpheme sequence, integrates the time relationship, and generates the speech text sequence; 语义分解模块对所述语音文本序列逐词分析词性关系，分离动词和目标对象，提取时间与范围条件，关联语义逻辑，生成任务语义描述集合；The semantic decomposition module analyzes the part-of-speech relationship of the speech text sequence word by word, separates verbs and target objects, extracts time and range conditions, associates semantic logic, and generates a task semantic description set; 任务分配模块解析所述任务语义描述集合中的关键任务和次要任务，分析资源和依赖顺序，优先排序分组并分配次序，分解链条，生成任务调度清单；The task allocation module parses the key tasks and secondary tasks in the task semantic description set, analyzes resources and dependency order, prioritizes and groups them, allocates them, decomposes the chain, and generates a task scheduling list; 逻辑执行模块加载所述任务调度清单中的任务属性，调用电子学生证检索课程，解析依赖并分类整合，执行查询任务，生成任务执行结果集合；The logic execution module loads the task attributes in the task scheduling list, calls the electronic student card to retrieve courses, parses dependencies and classifies and integrates them, executes query tasks, and generates a task execution result set; 动态响应模块基于所述任务执行结果集合分析已完成的查询内容，标记未完成任务，补充上下文为反馈文本，转化语音信号输出，获取用户反馈音频；The dynamic response module analyzes the completed query content based on the task execution result set, marks the unfinished tasks, supplements the context into feedback text, converts the voice signal output, and obtains the user feedback audio; 语音交互调整模块提取基于所述用户反馈音频提取对话内容与任务执行记录，分析未完成指令，调整任务链并分配流程，生成语音交互任务分配数据表。The voice interaction adjustment module extracts the conversation content and task execution records based on the user feedback audio, analyzes the unfinished instructions, adjusts the task chain and allocates the process, and generates a voice interaction task allocation data table. 2.根据权利要求1所述的智能语音交互系统，其特征在于：所述语音文本序列包括振幅特性、频率分布特性、语素序列、时间关系，所述任务语义描述集合包括时间限定条件、范围限定条件、逻辑关系、关键语义对象，所述任务调度清单包括关键任务、次要任务、任务优先级、任务依赖顺序、资源消耗评估，所述任务执行结果集合包括查询操作结果、任务分类数据、关联条件整合结果，所述反馈文本包括上下文信息、自然语言反馈内容、未完成任务标记，所述用户反馈音频包括查询内容、未完成指令、新指令内容，所述语音交互任务分配数据表包括未完成指令、重复调用记录、任务链条调整数据、调用流程分配信息。2. The intelligent voice interaction system according to claim 1 is characterized in that: the voice text sequence includes amplitude characteristics, frequency distribution characteristics, morpheme sequences, and time relationships; the task semantic description set includes time limitation conditions, scope limitation conditions, logical relationships, and key semantic objects; the task scheduling list includes key tasks, secondary tasks, task priorities, task dependency orders, and resource consumption assessments; the task execution result set includes query operation results, task classification data, and associated condition integration results; the feedback text includes context information, natural language feedback content, and unfinished task marks; the user feedback audio includes query content, unfinished instructions, and new instruction content; the voice interaction task allocation data table includes unfinished instructions, repeated call records, task chain adjustment data, and call process allocation information. 3.根据权利要求1所述的智能语音交互系统，其特征在于：所述语音输入处理模块包括：3. The intelligent voice interaction system according to claim 1, characterized in that: the voice input processing module comprises: 信号解析子模块通过解析语音信号的频域特征和时域特征，对信号的能量分布进行量化分析，对波形进行周期性检测和振幅分析，并调整频域分析的细节参数，得到信号特性描述；The signal analysis submodule analyzes the frequency domain and time domain characteristics of the speech signal, performs quantitative analysis on the energy distribution of the signal, performs periodicity detection and amplitude analysis on the waveform, and adjusts the detailed parameters of the frequency domain analysis to obtain a description of the signal characteristics; 帧分割子模块采用所述信号特性描述，进行音频信号的帧分割，调整分割的时间窗和间隔参数，优化帧内数据的连贯性和分辨率，提取每帧的振幅、频率和能量，通过帧内特性数据分析，获取帧特性数据集；The frame segmentation submodule uses the signal characteristic description to perform frame segmentation of the audio signal, adjust the time window and interval parameters of the segmentation, optimize the coherence and resolution of the data within the frame, extract the amplitude, frequency and energy of each frame, and obtain the frame characteristic data set through the analysis of the characteristic data within the frame; 语素整合子模块利用所述帧特性数据集，进行连续帧的语素分析和整合，通过时间序列分析强化语素间的逻辑和结构关系，优化语素的连贯性，生成语音文本序列。The morpheme integration submodule uses the frame characteristic data set to perform morpheme analysis and integration of continuous frames, strengthen the logical and structural relationship between morphemes through time series analysis, optimize the coherence of morphemes, and generate a speech text sequence. 4.根据权利要求1所述的智能语音交互系统，其特征在于：所述语义分解模块包括：4. The intelligent voice interaction system according to claim 1, characterized in that: the semantic decomposition module comprises: 词性分析子模块基于所述语音文本序列，进行词性标注，包括动词和名词的识别，使用语料库数据进行对比分析验证每个词的词性，调整分析精度并匹配文本复杂性，获取词性关系数据；The part-of-speech analysis submodule performs part-of-speech tagging based on the speech text sequence, including recognition of verbs and nouns, uses corpus data for comparative analysis to verify the part-of-speech of each word, adjusts the analysis accuracy and matches the text complexity, and obtains part-of-speech relationship data; 句法重组子模块采用所述词性关系数据，按照句法规则调整动词与目标对象之间的关系，优化句子结构，分析关键句法元素并重新组织语句成分，得到句法结构数据；The syntactic reorganization submodule uses the part-of-speech relationship data to adjust the relationship between the verb and the target object according to the syntactic rules, optimize the sentence structure, analyze the key syntactic elements and reorganize the sentence components to obtain syntactic structure data; 逻辑关系子模块基于所述句法结构数据，采用条件语义嵌套分析法，提取语句中的时间和范围限定条件，关联条件与语句关键语义对象，使用逻辑分析方法整合关键语义元素，生成任务语义描述集合。The logical relationship submodule is based on the syntactic structure data and adopts the conditional semantic nesting analysis method to extract the time and scope limiting conditions in the sentence, associate the conditions with the key semantic objects of the sentence, use the logical analysis method to integrate the key semantic elements, and generate a task semantic description set. 5.根据权利要求4所述的智能语音交互系统，其特征在于：所述采用条件语义嵌套分析法的公式如下：5. The intelligent voice interaction system according to claim 4, characterized in that: the formula using the conditional semantic nested analysis method is as follows: ； ; 其中，表示任务语义描述集合值，代表语句中时间限定条件的标准化值，代表语句中范围限定条件的语义密度，代表语句中关键语义对象的关联匹配度，代表逻辑分析过程中关键语义元素的权重总和，代表句法结构数据中层级的校正系数，代表语句中条件和关键语义对象的分布复杂性指数，、为调节系数，、为归一化系数。in, Represents the set value of the task semantic description, Represents the normalized value of the time qualification in the statement, represents the semantic density of the scope-qualifying conditions in the statement, Represents the correlation matching degree of key semantic objects in the sentence, Represents the sum of the weights of key semantic elements in the logical analysis process. represents the correction coefficient of the level in the syntactic structure data, represents the distribution complexity index of conditions and key semantic objects in the sentence, , is the adjustment coefficient, , is the normalization coefficient. 6.根据权利要求1所述的智能语音交互系统，其特征在于：所述任务分配模块包括：6. The intelligent voice interaction system according to claim 1, characterized in that: the task allocation module comprises: 任务分析子模块基于所述任务语义描述集合，识别和分类关键与非关键任务，对每项任务进行资源和时间消耗评估，优化资源分配和时间分配的精度，调整评估参数匹配差异化任务的紧急度，获取任务资源消耗数据；The task analysis submodule identifies and classifies critical and non-critical tasks based on the task semantic description set, evaluates the resource and time consumption of each task, optimizes the accuracy of resource allocation and time allocation, adjusts the evaluation parameters to match the urgency of differentiated tasks, and obtains task resource consumption data; 依赖排序子模块使用所述任务资源消耗数据，分析任务间的依赖关系和执行优先级，通过调整优先级排序参数优化任务执行顺序，生成任务依赖序列；The dependency sorting submodule uses the task resource consumption data to analyze the dependencies and execution priorities between tasks, optimizes the task execution order by adjusting the priority sorting parameters, and generates a task dependency sequence; 任务调度子模块依据所述任务依赖序列，制定任务执行分组和次序，规划每项任务的调度，调整执行分组参数和调度策略并匹配任务的需求，生成任务调度清单。The task scheduling submodule formulates the task execution grouping and order according to the task dependency sequence, plans the scheduling of each task, adjusts the execution grouping parameters and scheduling strategy and matches the task requirements, and generates a task scheduling list. 7.根据权利要求1所述的智能语音交互系统，其特征在于：所述逻辑执行模块包括：7. The intelligent voice interaction system according to claim 1, characterized in that: the logic execution module comprises: 任务属性加载子模块基于所述任务调度清单，识别每项任务的属性，对任务属性进行分析和加载，采用实时调整技术细化加载流程，调节加载速率和优化任务匹配逻辑，生成任务属性数据集；The task attribute loading submodule identifies the attributes of each task based on the task scheduling list, analyzes and loads the task attributes, refines the loading process using real-time adjustment technology, adjusts the loading rate and optimizes the task matching logic, and generates a task attribute data set; 课程信息检索子模块利用所述任务属性数据集，调用电子学生证数据进行目标课程信息的检索，包括调整和优化检索参数，提取关键课程依赖信息，获取课程依赖数据；The course information retrieval submodule uses the task attribute data set to call the electronic student ID data to retrieve the target course information, including adjusting and optimizing the retrieval parameters, extracting key course dependency information, and obtaining course dependency data; 依赖解析子模块通过所述课程依赖数据，整合关键任务和关联的条件任务，调整查询参数，并按顺序组织和执行，生成任务执行结果集合。The dependency parsing submodule integrates key tasks and associated conditional tasks through the course dependency data, adjusts query parameters, and organizes and executes them in sequence to generate a task execution result set. 8.根据权利要求1所述的智能语音交互系统，其特征在于：所述动态响应模块包括：8. The intelligent voice interaction system according to claim 1, characterized in that: the dynamic response module comprises: 查询结果分析子模块基于所述任务执行结果集合，进行查询结果分析，通过调节分析参数，区分已完成与未完成的查询任务，优化数据筛选逻辑，获取语音查询分析数据；The query result analysis submodule performs query result analysis based on the task execution result set, distinguishes completed and uncompleted query tasks by adjusting analysis parameters, optimizes data screening logic, and obtains voice query analysis data; 新指令标记子模块使用所述语音查询分析数据，对未完成的查询任务进行新指令标记，优化标记流程，强化识别的响应速度，调整分类参数，生成新指令任务集；The new instruction marking submodule uses the voice query analysis data to mark unfinished query tasks as new instructions, optimize the marking process, enhance the recognition response speed, adjust the classification parameters, and generate a new instruction task set; 自然语言反馈子模块依据所述新指令任务集，汇总针对用户的自然语言反馈，调整反馈参数，将反馈数据转换为语音信号，优化语音转换过程，捕获用户的反馈音频，得到用户反馈音频。The natural language feedback submodule summarizes the natural language feedback for the user according to the new instruction task set, adjusts the feedback parameters, converts the feedback data into a voice signal, optimizes the voice conversion process, captures the user's feedback audio, and obtains the user feedback audio. 9.根据权利要求1所述的智能语音交互系统，其特征在于：所述语音交互调整模块包括：9. The intelligent voice interaction system according to claim 1, characterized in that: the voice interaction adjustment module comprises: 对话分析子模块基于所述用户反馈音频，分析对话内容，包括与电子学生证任务执行记录关联的数据，通过优化语音到文本的转换精度，区分和识别未完成和重复指令，强化对细节处理的分析，生成细节对话分析结果；The conversation analysis submodule analyzes the conversation content, including data associated with the electronic student ID card task execution record, based on the user feedback audio, and generates a detailed conversation analysis result by optimizing the voice-to-text conversion accuracy, distinguishing and identifying unfinished and repeated instructions, and strengthening the analysis of detail processing; 任务链重组子模块使用所述细节对话分析结果，重新排序对话中识别的任务，调整任务的执行顺序并反映实时用户需求的优先级，微调任务管理逻辑优化执行路径，获取重组任务链数据；The task chain reorganization submodule uses the detailed conversation analysis results to reorder the tasks identified in the conversation, adjust the execution order of the tasks and reflect the priority of real-time user needs, fine-tune the task management logic to optimize the execution path, and obtain the reorganized task chain data; 流程重新分配子模块根据所述重组任务链数据，进行电子学生证功能的调用流程重新分配，调整任务的调用顺序和资源分配，匹配用户交互需求，优化调用流程，生成语音交互任务分配数据表。The process reallocation submodule reallocates the calling process of the electronic student card function according to the reorganized task chain data, adjusts the calling sequence and resource allocation of tasks, matches user interaction requirements, optimizes the calling process, and generates a voice interaction task allocation data table. 10.一种电子学生证，包括存储器和处理器，其特征在于，所述存储器中存储有计算机程序，所述处理器执行所述计算机程序时实现权利要求1至9任一项所述的智能语音交互系统。10. An electronic student ID card, comprising a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the intelligent voice interaction system according to any one of claims 1 to 9 when executing the computer program.
CN202510006424.7A 2025-01-03 2025-01-03 An intelligent voice interaction system and electronic student ID card Active CN119418699B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202510006424.7A CN119418699B (en) 2025-01-03 2025-01-03 An intelligent voice interaction system and electronic student ID card

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202510006424.7A CN119418699B (en) 2025-01-03 2025-01-03 An intelligent voice interaction system and electronic student ID card

Publications (2)

Publication Number Publication Date
CN119418699A CN119418699A (en) 2025-02-11
CN119418699B true CN119418699B (en) 2025-03-11

Family

ID=94462009

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202510006424.7A Active CN119418699B (en) 2025-01-03 2025-01-03 An intelligent voice interaction system and electronic student ID card

Country Status (1)

Country Link
CN (1) CN119418699B (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN119597937B (en) * 2025-02-12 2025-07-11 南通市志达通信息服务有限公司 Intelligent outbound knowledge graph construction method and system based on dialogue scene
CN120146064B (en) * 2025-03-06 2026-01-23 中原环保股份有限公司 A digital human interaction management system based on big data analysis
CN120218538A (en) * 2025-03-21 2025-06-27 云南航信空港网络有限公司 A method based on the mission warning model of the entire process of aircraft support
CN120431919A (en) * 2025-04-03 2025-08-05 北京铁力山科技股份有限公司 An artificial intelligence-based voice interactive console
CN120447421B (en) * 2025-07-14 2025-09-19 湖南大学 A family service method and system based on artificial intelligence

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6665640B1 (en) * 1999-11-12 2003-12-16 Phoenix Solutions, Inc. Interactive speech based learning/training system formulating search queries based on natural language parsing of recognized user queries
WO2015036817A1 (en) * 2013-09-15 2015-03-19 Yogesh Chunilal Rathod Structured updated status, requests, user data & programming based presenting & accessing of connections
WO2015113578A1 (en) * 2014-01-28 2015-08-06 Luciano Zorzin Method for automatically detecting meaning and measuring the clearness of text
US20180061256A1 (en) * 2016-01-25 2018-03-01 Wespeke, Inc. Automated digital media content extraction for digital lesson generation

Also Published As

Publication number Publication date
CN119418699A (en) 2025-02-11

Similar Documents

Publication Publication Date Title
CN119418699B (en) An intelligent voice interaction system and electronic student ID card
KR102897602B1 (en) Machine learning system for digital assistants
CA2437620C (en) Hierarchichal language models
US8515736B1 (en) Training call routing applications by reusing semantically-labeled data collected for prior applications
CN120015053B (en) AI-driven personalized voice training and pronunciation correction system
CN119357069B (en) Automatic testing method for language ability of intelligent body based on multi-language model self-evolution
CN119887139A (en) Intelligent information screening method and system for human resource system
Cao et al. Speaker-independent speech emotion recognition based on random forest feature selection algorithm
CN120123474A (en) An AIGC empowerment assistance system and implementation method for second foreign language teaching in universities
CN117352000A (en) Speech classification method, device, electronic equipment and computer readable medium
CN120672104A (en) Dispute reconciliation key information extraction system based on user feedback
CN120929616B (en) Multi-mode heterogeneous model retrieval enhancement method and system
CN117809655A (en) Audio processing method, device, equipment and storage medium
CN113822506A (en) A multi-round voice interactive intelligent retrieval system and method for power control
Hu et al. Automatic analysis of speech prosody in Dutch
CN118170919B (en) A method and system for classifying literary works
CN117857599B (en) Digital person dialogue intelligent management system based on Internet of things
CN118069805A (en) Intelligent question-answering method and device based on voice and text collaboration
Holmström et al. Effect of feature extraction when classifying emotions in speech-an applied study
Rizvi et al. Unveiling emotions: Long-short term for speech emotion recognition
Bertero et al. Towards Universal End-to-End Affect Recognition from Multilingual Speech by ConvNets
Vaidehi et al. A Machine Learning and Deep Learning based Approach to Generate a Speech Emotion Recognition System
CN120597885A (en) Intention recognition processing method, device and terminal based on learning machine
CN121391198A (en) Full-flow automatic interview assisting method, system and equipment based on large language model
CN120727012A (en) Intelligent voice interaction method and device based on large language model

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant