469 results about "Language recognition" patented technology

The invention discloses a sign language translation method and device based on vision and word feature pre-training alignment, and belongs to the technical field of sign language translation.The sign language translation method comprises the steps that vision features of sign language videos are extracted through a vision encoder, and word text features are extracted through a text encoder; performing comparative learning based on vision and word text features to obtain a pre-trained vision encoder; performing word text mask prediction pre-training on the text decoder; constructing the pre-trained visual encoder and the pre-trained text decoder into a sign language recognition model so as to recognize a word text sequence from the sign language video; and accessing the pre-trained large language model in the domain into a sign language recognition model to construct a sign language translation model, and performing combined fine tuning so as to translate the word text sequence into a natural language text. The sign language recognition and translation method can realize more efficient, accurate and reliable sign language recognition and translation, is applied to the fields of intelligent sign language translation, barrier-free communication, sign language education and the like, and provides more accurate and natural language interaction experience for hearing-impaired groups.

The invention relates to a model for evaluating and predicting mild cognitive impairment risk of old people in a pension institution. The model sequentially comprises a behavior analysis module, a language recognition module, a social modeling module, a toughness calculation module, a feature fusion module, a risk reasoning module and the like. Behavior deviation characteristics and abnormal time periods are extracted by collecting behavior data of daily life, diet, social contact and the like of old people and comparing the behavior data with an institution work and rest template; in combination with nursing records, extracting language anomaly features; analyzing social frequency and structure changes in the abnormal time period, and extracting social variation features; a cognitive toughness index is calculated by integrating the health archive and the recovery ability to the health event; and performing toughness weighting on the multi-dimensional features to construct a time sequence tensor, and inputting the time sequence tensor into a recursive model to predict a cognitive impairment risk value. And if the risk value suddenly changes, the system automatically backtracks the feature trajectory of nearly 7 days, constructs and screens a prediction path with the strongest interpretation force, outputs a dominant prediction result and a key factor sequence, and realizes high-interpretability and high-reliability early recognition and intervention reference.

The invention provides an AI-assisted large-model intelligent question and answer accuracy improvement method, and belongs to the technical field of intelligent question and answer. Comprising the steps of data collection, data cleaning, data labeling, dynamic knowledge base construction, knowledge base updating trigger mechanism establishment, pre-training language model improvement, post-processing language understanding enhancement, combination of rule-based and machine learning reasoning, reasoning result verification and optimization, dialogue strategy optimization, emotion perception and response, multi-language search and language recognition search. And multi-dimensional evaluation index setting and regular evaluation and feedback improvement are realized. The intelligent question and answer accuracy improving method is high in understanding ability, timely in data updating, outstanding in reasoning ability and high in language recognition and dialogue ability.

The invention discloses a continuous sign language recognition method based on multi-granularity cross-modal comparative learning. The method comprises the following steps: firstly, fully tapping a hidden semantic mapping relationship between a hidden sign language video and a corresponding sign language text; then, a multi-granularity cross-modal comparison mechanism is used for relieving potential influences of redundant frames and semantic irrelevant identification words on cross-modal alignment, and through a coarse-granularity video-sentence comparison strategy and a fine-granularity frame-word comparison strategy, visual and text modal features are effectively aligned from global and local angles; semantic interference caused by redundant frames and semantic irrelevant identification words is relieved through cross-granularity video-word comparison and cross-granularity frame-sentence comparison strategies; and finally, obtaining sign language text sentences through a decoder by utilizing visual embedding obtained in a visual encoder, and taking the sign language text sentences as predicted continuous sign language recognition results. The hidden semantic mapping relation between the sign language video and the corresponding sign language text is fully excavated, the semantic understanding of the sign language model on the visual context is enhanced, and the applicability and accuracy of sign language recognition are improved.

The invention relates to the technical field of language recognition, in particular to a multi-dialect voice text conversion system and processing method, and the system comprises an audio collection module which is responsible for collecting voice input of a user and supports PCM, MP3 and WAV formats and sampling rates of 16kHz and 44.1 kHz; the dialect detection module is used for recognizing a dialect type of voice input by utilizing a pre-training language classifier; the voice feature extraction module is used for converting the voice signal into feature representation, such as Mel spectrum and MFCC (Mel Frequency Cepstrum Coefficient); the multi-dialect speech recognition module is used for mapping the features into a text sequence based on a speech recognition model of multi-task learning; and the text post-processing module is used for outputting a final standardized text according to a language rule and a context optimization recognition result. The multi-dialect support covers main dialect languages and can be quickly expanded to new dialects through transfer learning.

The invention discloses a sign language recognition system based on a dynamic time warping algorithm, and the system comprises a data collection module which is used for collecting gesture data in real time through a camera and a depth sensor; the feature extraction and time sequence conversion module is used for converting the gesture data into a time sequence; the dynamic time warping matching module is used for matching the input gesture with the gesture in the template library by utilizing a dynamic time warping algorithm; the matching result selection and template classification module is used for selecting the most matched template according to the similarity and determining the gesture category; the embedded platform and hardware acceleration module is used for executing a sign language recognition algorithm on the embedded platform and improving the recognition speed and real-time performance through hardware acceleration; and the character and voice output and real-time feedback optimization module is used for converting the recognition result into characters and voice output. Through hardware acceleration and data compression optimization, the sign language recognition precision, real-time performance and adaptability are improved, and the sign language recognition method is suitable for an embedded system with limited resources.

The invention discloses a skeleton sign language recognition method of a double-flow space-time dynamic graph convolutional network fused with residual learning, and belongs to the technical field of artificial intelligence and gesture recognition. According to the method, an input gesture skeleton sequence relative to a face is divided into two data streams, namely a hand form data stream and a wrist track data stream through double-reference-system differential homeomorphic mapping; the method comprises the following steps: firstly, processing hand posture data, and capturing a hand joint spatial topological relation by combining a spatial-temporal dynamic graph convolutional network (STDGCNN) with a residual convolutional block; meanwhile, a Finsler trajectory dynamics encoder (FTDE) is adopted to carry out differential geometric modeling on the wrist trajectory, and the direction sensitivity characteristic of the trajectory is captured through a multi-scale causal convolutional network. Then, mutual enhancement of double-flow features is realized through a bidirectional cross feature enhancer (BCFE), and the problem of geometric inconsistency of a heterogeneous feature space is solved through a geometric-driven optimal transmission fusion device (Geometric-OT). The method solves the challenge that a traditional sign language recognition method processes complex space-time correlation of gesture forms and motion tracks at the same time, the technical problems of insufficient relation between hands and faces, insufficient feature expression ability and low space-time feature extraction efficiency, and the problems of geometric inconsistency, single reference system and the like. And the identification accuracy and the real-time performance are obviously improved. Experiments show that the accuracy rate of the method in complex hundreds of sign language vocabulary recognition tasks reaches 95% or above on average, the reasoning speed is only 17ms on average, high-precision real-time sign language recognition is achieved, and the method has higher robustness in complex environments such as noise and shielding.

The invention discloses a sign language recognition method and device based on multi-modal deep learning, and the method comprises the steps: multi-modal data input: capturing hand motions, gesture tracks and facial expressions at the same time through a camera, a motion capture sensor and other devices, and forming multi-modal data input; sign language action recognition: precise recognition of sign language actions is realized through a combined model of a deep convolutional neural network and a long-short-term memory network; facial expression and gesture track combined recognition: realizing understanding and translation of complex sign language sentences by combining the captured facial expressions and gesture tracks; and context natural language processing: generating a target statement in combination with context semantic understanding, and outputting a translation result. Through hand motion capture, facial expression analysis, gesture trajectory tracking and context natural language processing, complex sign language motions can be recognized more accurately and translated into characters or voices in real time, and low delay and high accuracy are achieved.

The invention belongs to a decoding method, and provides a multi-modal brain-computer interface decoding method and a related device for solving the technical problems that an existing non-intrusive brain language decoding method is insufficient in adaptability in global context and weak in generalization performance in a cross-subject scene, and multi-modal neural feature collaborative enhancement is difficult to achieve. And determining the called execution module. The execution module comprises a feature extraction module, a cross-subject standardization module, a multi-mode semantic fusion module, a language recognition module and a semantic consistency module. By obtaining the unified semantic representation and combining the beam search algorithm, the fairness and universality in different language groups are remarkably improved, multiple modes can be supported, and the brain signal decoding precision and robustness are improved. In addition, cross-subject semantic representation generalization can be realized, and individual specificity is effectively reduced.

The invention discloses an artificial intelligence automatic response conference assistant, and particularly relates to the technical field of intelligent video conferences. The system comprises a user interface used for displaying conference information, function options and execution results, a large language model used for natural language processing and understanding, and a language recognition module used for converting voice input into characters to achieve real-time conference recording and voice instruction recognition. By means of friendly user interface design, conference information, function options and execution results can be visually displayed, a user can easily operate and master the whole conference process, the voice recognition module adopts a high-quality recording module, voice content in a conference is accurately transferred into characters in real time, and the user experience is improved. The completeness and the accuracy of conference records are ensured, the large language model further performs semantic understanding and processing on the transferred text, intelligent conference abstracts and responses are generated, a user is helped to rapidly obtain key information, and the postprocessing module automatically sends the abstracts to a user mailbox to achieve the purpose of meeting. Or the system is automatically integrated to an online office system of a user, so that a lot of time and energy are saved.

The invention discloses an intelligent cockpit implementation method and system based on single-model multi-task reasoning, and belongs to the field of automotive electronics, and the method comprises the steps: inputting shared depth features into a detection head and a first classification head after original image data in an intelligent cockpit is preprocessed; the output of the detection head is subjected to detection post-processing to obtain a target detection anchor frame and position information, and the output of the first classification head is subjected to classification post-processing to obtain distraction identification classification; after the target detection anchor frame and the position information are preprocessed, human body area sub-images are input into a safety belt recognition and classification model; inputting the face region sub-images into a fatigue recognition classification model, a face ID matching model, a fixation point estimation model, a sight line estimation model and a head posture estimation model in parallel; and a face image sequence and a voice sequence in the continuous video stream are collected, the face image sequence and the voice sequence are respectively input into the lip language recognition model and the voice recognition model, a recognition result is input into the large language model to be processed to generate an interaction instruction, and a response is output through the voice synthesis module.

The invention discloses a neck-hung sign language interpretation device and a sign language semantic recognition interpretation method thereof.The neck-hung sign language interpretation device comprises a wearing support which defines a neck surrounding space and comprises a middle connecting arm and two shells, and the two shells are connected to the two sides of the middle connecting arm respectively; the image acquisition unit is arranged on the front side of the shell, comprises a three-dimensional depth camera and an adjusting mechanism for driving the three-dimensional depth camera to rotate, and is used for dynamically tracking and acquiring continuous sign language videos; the processing module is integrated in the shell, comprises a main control processor and a memory, and is used for operating a sign language recognition algorithm; the voice output unit is electrically connected with the main control processor and is used for playing the translation result; and the power supply unit supplies power to the image acquisition unit, the processing module and the voice output unit. Gesture languages used by the deaf-mute for communication are converted into natural languages which can be easily understood by normal people, the deaf-mute group is helped to speak, and the problem of information output when the deaf-mute communicates with the normal people is solved.

The invention is suitable for the technical field of intelligent voice processing, and provides a language classification model training method, a language recognition method and device and intelligent equipment. The language classification model training method comprises the following steps: acquiring a first acoustic feature, wherein the first acoustic feature refers to an acoustic feature of an audio clip; the acoustic features of the audio clips are sequentially input into N layers of encoders of an initial language classification model, first encoding features output by the N layers of encoders respectively are obtained, and N is an integer larger than 1; inputting each first coding feature into a fusion module of the initial language classification model to obtain a first fusion feature; and based on the first fusion feature and a language recognition module of the initial language classification model, training the initial language classification model to obtain a target language classification model, the target language classification model being used for recognizing a language category of the audio data. According to the invention, the accuracy and robustness of language recognition can be improved.

The invention relates to the technical field of voice signal processing and synthesis, and particularly discloses an information processing method and system for voice conversion, and the method comprises the steps: extracting a language feature vector of each word from an input text; calculating posterior probabilities of the words in different languages by combining a Bayesian reasoning mechanism, and generating language attribution confidence coefficient characteristic values; a Monte Carlo sampling method is adopted to carry out multiple times of context sensitive simulation, and pronunciation path selection probability distribution characteristic values are generated; further fusing the feature values into a multi-language pronunciation decision vector, inputting the multi-language pronunciation decision vector into a multi-language end-to-end speech synthesis model, calling a phoneme mapping rule of a corresponding language and an acoustic parameter prediction module, and generating a high-quality target speech spectrogram; and finally, dynamically adjusting Bayesian prior distribution and a language recognition threshold according to the output speech spectrogram.

The invention relates to a sign language action recognition problem, and provides a sign language recognition method based on a millimeter wave radar 3D point cloud, sign language recognition is carried out in a sequence-to-sequence mode through a non-contact FMCW radar, and the sign language recognition method mainly comprises a first part, namely a continuous sign language action recognition method based on a millimeter wave radar; a second part: establishing an evaluation index STT Score of radar continuous sign language action recognition; according to the invention, a low-cost portable wearable acquisition device is adopted, so that a subject can use the device anytime and anywhere; meanwhile, millimeter-wave radar detection does not depend on light rays, the calculated amount is small, the privacy of a user is effectively protected due to the fact that signals collected by the millimeter-wave radar are radar radio frequency signals, and meanwhile the use feeling is good and the cost is low due to the non-contact collection mode of the millimeter-wave radar.

According to the method and device for generating the fine-grained semantic description from the action video data, the training data set is established based on the isolated word sign language recognition data set and the continuous sign language recognition data set containing the word target notation, and the action video data and the action description text data of fine-grained semantic description modeling are obtained; a fine-grained semantic action description style pre-training generation model is obtained through a training framework comprising an action video feature coding module, a multi-modal feature fusion module and a text feature coding module by combining user cue words and system cue words and introducing a mask reconstruction mechanism, action video data is adopted for fine tuning, a loss function is established, and a fine-grained semantic action description style is obtained. The fine-grained semantic action description generation model is used for generating high-quality fine-grained semantic action description data, and the problem that the current fine-grained semantic action description data is insufficient is solved. And the stability and the accuracy of a generated result are ensured when high-dynamic complex scenes such as sign language videos and interactive actions are processed.

The embodiment of the invention discloses a multi-language text transcription method and system, and is applied to an AR device end, the method comprises the following steps: collecting first to-be-transcribed data of a user through the AR device end, and preprocessing the first to-be-transcribed data to obtain preprocessed to-be-transcribed data, the first to-be-transcribed data comprising voice data, visual data and action data; performing feature extraction on the first to-be-transcribed data to obtain first to-be-transcribed feature data, the first to-be-transcribed feature data including first voice feature data, first visual feature data and first action feature data; inputting the first voice feature data into a preset language recognition model, outputting preliminary target language data, and sending the first to-be-transcribed feature data and the preliminary target language data to an edge device end; and receiving the text transcription result, and displaying the text transcription result on the AR equipment end, so that the problem of insufficient computing resources of the AR equipment end can be effectively solved, and the transcription accuracy and reaction speed of the AR equipment end are improved.

The invention discloses a sign language recognition glasses device, system and method, and belongs to the technical field of intelligent equipment, and the sign language recognition glasses device comprises a glasses frame which is used for supporting all components of the device; the camera is arranged on the front side of the glasses frame and is used for capturing hand actions in real time to generate image information; the data processing module is embedded in the glasses frame, electrically connected to the camera and used for processing the image information and executing sign language recognition; the main control unit is arranged on an ear rod of the glasses frame, is electrically connected to the data processing module and is used for coordinating the operation of each component; the display module is arranged at the positions of the lenses of the glasses frame, electrically connected to the main control unit and used for displaying character information obtained after sign language recognition, communication between hearing-impaired people and common people is not limited by specific places any more, and instantaneity and convenience of communication are greatly improved.

The invention discloses an artificial intelligence-assisted cross-language automatic note taking and term tagging system, and belongs to the technical field of language tagging. The system firstly obtains and processes audio data, generates a preliminary text through cross-language speech recognition, recognizes term candidate words in the text, and performs pronunciation association matching with a term knowledge base; and performing language recognition and context analysis on the candidate words and the contexts thereof, performing comprehensive pronunciation matching, language recognition and context score calculation to obtain a comprehensive score, thereby accurately selecting a cross-language term which most conforms to the current context as the final candidate word, and obtaining the annotation information of the final candidate word. And integrating the final candidate word and the annotation information of the final candidate word into the calibration note text, and outputting an automatic note with cross-language term annotations. According to the method, complex audios including language mixing and the like can be automatically processed, the high-quality calibration note text is generated, and the information recording efficiency and accuracy in a cross-language communication scene are remarkably improved.

The invention relates to the technical field of sign language intelligent communication, in particular to a deaf-mute sign language translation pronunciation system which comprises a data acquisition and preprocessing module, a sign language model recognition module, a dynamic gesture model module and a voice conversion module. According to the sign language translation pronunciation system for the deaf-mute, wide-angle camera intelligent glasses are adopted for visual collection, wearable equipment is not needed, video streams and thermodynamic diagrams are combined, visual information and key point probability distribution are complementary, the influence of single-mode noise is reduced, video space-time features are extracted through a 3D CNN, long-range dependence is captured in combination with a self-attention mechanism of Transform, and therefore the sign language translation pronunciation system for the deaf-mute is obtained. The sign language recognition precision is improved, an internal reference matrix and a distortion coefficient are calculated through a calibration board, wide-angle lens distortion is corrected, it is ensured that hand key points are accurately positioned, YOLOv8 is adopted to segment an interference object, background noise is prevented from affecting recognition, 21 key point thermodynamic diagrams are generated through MediaPipe, fault tolerance of low-confidence-coefficient key points is enhanced through Gaussian kernel diffusion, and the recognition accuracy is improved. And the model identification precision is improved.

The invention discloses a continuous sign language recognition method based on layered space-time enhancement. The method comprises the following steps: acquiring a sign language video; and inputting the sign language video into the trained sign language recognition model to obtain a first recognition result and a second recognition result, and taking the first recognition result as a final sign language recognition result. According to the continuous sign language recognition method based on hierarchical space-time enhancement, multi-stage output of a ResNet34 network is captured through an alignment module, and additional hierarchical alignment supervision is provided; according to the method, the alignment module and the timing causal module are integrated into the ResNet34 network, future and past information is aggregated through the timing causal module, so that more accurate vocabulary boundary perception is achieved, the alignment module and the timing causal module are integrated into the ResNet34 network, good balance between accuracy and calculation cost is achieved, an efficient and accurate solution is provided for a continuous sign language recognition task, and the continuous sign language recognition efficiency is improved. The recognition performance and robustness are remarkably improved, and the problem of under-fitting of the ResNet34 network in space and time contexts is solved.

The invention provides a vehicle-mounted voice recognition method and device, electronic equipment and a storage medium, and belongs to the technical field of voice recognition, and the method comprises the steps: carrying out the fusion of a voice recognition result and a lip language recognition result of a cab, obtaining a multi-modal recognition result, enabling the voice recognition result to be obtained through the recognition of a voice signal collected in real time, and enabling the lip language recognition result to be a lip language recognition result; the lip language recognition result is obtained by recognizing a mouth shape image collected in real time; analyzing the multi-modal recognition result to obtain a voice source of each voice and an intention corresponding to each voice; and based on the voice source and intention of each voice, respectively generating a vehicle control instruction. According to the invention, the voice recognition result and the lip language recognition result are combined to distinguish the voice from different sources, so that the main driver and the co-driver can respond to the voice instructions when sending the voice instructions at the same time, and the requirements of the main driver and the co-driver can be met at the same time.

The invention belongs to the field of millimeter-wave radar sensing and hand posture estimation, and discloses a millimeter-wave radar-based multi-target hand posture estimation method, which comprises the following steps of: 1, creating a hand region: detecting all objects in a millimeter-wave radar sensing region, detecting a predefined wake-up gesture through a wake-up detection model to recognize a target user, the method comprises the following steps: step 1, evaluating the occurrence probability of a wake-up gesture, and recognizing a hand region, step 2, in the created hand region, constructing a distance-azimuth-pitching three-dimensional feature cube which represents hand posture features; and step 3, estimating hand posture features through an m < 2 > HandNet model. The millimeter wave radar is used as a research object, hand posture estimation is achieved through the millimeter wave radar, multi-user and multi-hand interaction is supported, the shielding problem is effectively solved, and wide application scenes such as intelligent home control, intelligent terminal interaction, virtual reality equipment and sign language recognition are met.

The invention relates to the technical field of software engineering, and discloses an assembly line flow adaptive configuration method and system. The method comprises the following steps: obtaining source code data of a software project and performing static analysis and feature extraction to generate structured language description data; querying and matching in a preset template library based on the structured language description data to generate a task template set; performing task dependency relationship analysis and process topology assembly on the task template set to generate initial pipeline data; and performing dynamic strategy adjustment on the execution process of the initial pipeline data by adopting the real-time context data corresponding to the software project to generate target pipeline data. Through automatic language recognition and intelligent process generation, automatic configuration of the continuous integration process is realized, and the technical problem of low automation level caused by lack of automatic recognition and adaptation capability for a project development language environment in a project continuous integration and continuous deployment technology is solved.

An intelligent body system for gesture recognition and silent language recognition comprises a signal acquisition module composed of flexible strain sensors, a signal processing module, a language enhancement module, an augmented reality interaction module and an intelligent auxiliary module. Gestures or lip actions input by a user are recognized, after preliminary classification is conducted through a neural network, a large language model is called to achieve language completion and multi-language translation, an augmented reality interface is used for displaying interaction content, meanwhile, intelligent home equipment can be controlled, and the control state can be fed back in real time. The system is particularly suitable for communication of groups with sound production difficulty, has the advantages of humanization, high recognition rate, strong language complementation ability, natural interaction mode and the like, and realizes a complete intelligent interaction closed loop from perception to understanding and from recognition to control.

The invention relates to a sign language-to-multilingual text speech mutual conversion method based on deep learning and application thereof, belongs to the field of artificial intelligence, man-machine interaction and barrier-free communication, and particularly relates to a sign language recognition, semantic understanding and multilingual speech synthesis method based on deep learning and an application system thereof. The sign language to voice text conversion method comprises the following steps: 1, an image acquisition and preprocessing module, 2, a hand key point detection module, 3, time sequence modeling and sign language recognition, 4, natural language understanding and semantic optimization, 5, a multi-language translation module, and 6, a text and voice synthesis output module. According to the sign language-to-multilingual text voice mutual conversion method based on deep learning and the application thereof, sign languages can be automatically, flexibly and accurately converted into voices and characters, and deaf-mutes are helped to efficiently communicate with common people; the system can be widely applied to hospitals, schools, government affair windows, public transportation and other scenes.

The invention relates to the technical field of medical software, in particular to a system and method capable of recognizing early Alzheimer's disease through mobile phone voice, the system capable of recognizing early Alzheimer's disease through mobile phone voice comprises intelligent mobile phone software and a hospital background, and the mobile phone software builds an interactive platform based on the hospital background. Through intervention monitoring of mobile phone software, when voice communication, chatting and instruction input are carried out in daily life, monitoring can be carried out through system monitoring of software and characteristics such as smoothness and continuity of voice, so that analysis and diagnosis of the Alzheimer's disease can be carried out according to comparison with a normal state, and moreover, in the monitoring process, the monitoring efficiency is improved. Diversified collection is carried out in different time periods and different statement numbers, language recognition and dialect recognition are added when a user inputs voice, and keyword recognition for determining the Alzheimer's disease condition can be carried out through approximate comparison with mandarin after the user speaks out statements.

The invention discloses a language recognition method and system based on an artificial intelligence auxiliary conversation scene, and belongs to the technical field of language recognition, and the method specifically comprises the steps: collecting an audio signal and video data through a sensor, carrying out the noise filtering of the audio signal and the video data in real time, carrying out the preprocessing of the audio signal after noise filtering, and carrying out the recognition of the voice. Separating voices in the preprocessed audio signals, segmenting talkers, extracting voice features of each talker, performing voice decoding according to the voice features, outputting a final text, and analyzing intentions of the talkers according to the final text and context output; facial expressions and body languages of the talker are captured through the camera, and the real semantics of the dialogue can be more comprehensively understood in combination with tone and emotion information in the voice signal.

The invention discloses a lip language recognition method and device based on a space-time network model, a medium and a product, and relates to the technical field of lip language recognition. The method comprises the following steps: acquiring a to-be-recognized lip language video image; identifying a lip position in each frame of the to-be-identified lip language video image, and segmenting the corresponding lip image according to the lip position to obtain a plurality of target lip images arranged according to a time sequence; inputting the plurality of target lip images into a lip language recognition model to obtain a lip language prediction result; the lip language recognition model comprises a space-time network model, a gating circulation unit, an attention mechanism and a softmax function; the lip language prediction result is a character expressed by the to-be-recognized lip language video image. The method can improve the recognition precision and detection speed of the lip language.

The invention provides a microstack-based reasoning language model training and reasoning method, a model and computer equipment, and the method comprises the steps: calling a to-be-trained model to sequentially carry out multiple times of feature extraction on initial semantic features of a language, and obtaining predicted semantic features; calculating a loss value between the predicted semantic feature and a standard semantic feature of the language; and if the loss value satisfies a preset convergence condition, determining the to-be-trained model as a target inference language model based on a microstack, the target inference language model based on the microstack having a function of identifying semantic features of the language. The to-be-trained model comprises i feature extraction layers, and when the feature extraction layers extract features, data stored in the preset storage unit is added to input data on the basis that only the output data of the previous feature extraction layer exists, so that the accuracy of language recognition is improved.