820 results about "Voice pitch" patented technology

An AI-driven system for personalized advertising in real time, where: ◯ an analytics unit to monitor and analyze user behavior in real time across multiple digital platforms such as websites, mobile applications, social media, and smart devices; the unit collects data on user engagement, browsing patterns, time spent, and content preferences to enable targeted, personalized advertising; ◯ a prediction module that predicts preferences and interests of a user, operatively connected to the user interaction data collection unit, wherein the prediction module uses machine learning models such as deep learning, recurrent neural networks (RNNs) and transformer-based architectures to predict interests of users based on historical interactions and inferred preferences; ◯ an emotion and sentiment analysis unit that assesses the user's mood in real time through computer vision, natural language processing (NLP) and voice analysis, whereby the analysis of facial expressions, voice pitch and linguistic mood is used to determine emotional states and receptivity to advertising content; ◯ an embodiment of a context awareness component in operational communication with the emotion analysis and mood unit, in some cases further augmented by various environmental and situational data such as the device type and its physical location, date and time, and the content processed in the device, processing methods, etc., in establishing adaptability and automatic ad placement to be as relevant as possible to the user and their status as prescribed; ◯ Use reinforcement learning algorithms and generative AI models to drive advertising with personalization engines. Creative elements, messaging, and presentations are dynamically adjusted in real time based on user responses to ensure advertising is personalized and always optimized for best performance; o a privacy-focused AI system with federated learning, differential privacy methods, and on-device AI processing to reduce targeted advertising while complying with international data protection laws such as the General Data Protection Regulation (GDPR) and California consumer privacy laws; o an operationally adapted ad delivery mechanism to engage with real-time bidding (RTB) networks, programmatic advertising exchanges and demand-side platforms (DSPs) and place advertisements through digital advertising networks, which guarantees the delivery of tailored advertising to the most relevant audience in real time; and o a contextual feedback loop in which the machine learning models used in the preference and interest prediction module and in advertising personalization The engine is continuously updated to reflect the latest user engagement data, improving personalization over time and optimizing advertising performance.

The invention provides an AI-based virtual digital human interaction system, and relates to the technical field of data processing, and the system is used for obtaining a user feature data set by performing feature extraction on an interaction data set, calculating an audio change rate and speech speed stability, obtaining speech emotion data, calculating syntactic complexity and counting the number of emotion words. Calculating a facial expression symmetry degree; obtaining facial emotion data; calculating body stability and identifying a visual staying area; obtaining attention state data; adjusting tone and speed of the virtual digital person according to the virtual digital person text, optimizing sentence pattern complexity and sentiment words in the virtual digital person text, adjusting changes of eyes, eyebrows and mouth corners of the virtual digital person, and adjusting information display sequence and visual focus guidance. According to the method, the interaction efficiency of the virtual digital human is improved by analyzing the facial expressions and the body actions of the user.

The invention discloses an emotion prediction and disease derivation method and system based on multi-modal fusion, and the system comprises a data collection and preprocessing module, an emotion fusion module, an abnormal condition detection and cloud uploading module, and a disease possibility derivation module. The data acquisition and preprocessing module comprises a video part, a text part and an audio part, and the video part comprises face emotion recognition and prediction and human motion recognition and prediction; the text part comprises text content emotion recognition and prediction; the audio part comprises voice-to-text and voice tone emotion recognition and prediction, the system comprehensively captures an emotion state by fusing multi-mode information such as video, text and voice, and the accuracy and prediction capability of emotion recognition are improved; and by predicting the future emotion trend, the abnormal condition is warned in advance, and the response timeliness is improved.

The invention relates to the technical field of data processing, and particularly discloses a digital human voice generation system based on a large language model, which is used for solving the existing problems and comprises a semantic analysis module, a voice generation module, a multi-mode synchronization module, a data transmission module and a user feedback module, the semantic analysis module performs semantic analysis and sentiment analysis on voice and text data input by a user by using a large language model, extracts voice information and sentiment metadata for describing context, and dynamically constructs a semantic structure and a sentiment label adaptive to an education scene according to input content based on a metadata automatic derivation technology; through the dynamic time warping algorithm and the educational knowledge graph, semantic and emotion changes are accurately captured, the speech speed and tone are flexibly adjusted, and the naturalness of digital human speech generation and the adaptability of educational scenes are improved.

A system for AI-based analysis of student behavior, consisting of: a variety of data collection devices, including cameras, microphones and wearable sensors; a central processing unit (CPU) that is communicatively connected to the data acquisition devices; and a behavioral analysis engine, wherein: (a) the cameras are configured to capture visual data indicating student behavior, including facial expressions, body language and eye movements, with the captured visual data being processed using computer vision techniques to detect certain emotional and engagement-related indicators such as stress, boredom, attention or excitement; (b) the microphones are configured to capture audio data, including speech patterns, tone of voice, volume and level of participation, the captured audio data being processed using natural language processing (NLP) techniques to assess the mood, engagement and verbal participation of students in classroom activities; (c) the wearable sensors are configured to collect biometric data, including heart rate, skin temperature and movement, with the biometric data being analyzed to detect physiological signs of emotional stress, fatigue or engagement; (d) The behavioral analysis engine is configured to process the captured visual, auditory, and biometric data in real time to create a behavioral profile of each student. It uses machine learning models to correlate certain patterns with academic performance, emotional well-being, and social engagement. (e) the system is also configured to create predictive models based on historical data to identify pupils at risk of declining performance, social isolation or emotional distress, and to send alerts to teachers, counselors or the administration to enable timely intervention to address the identified problems.

The invention discloses a voice cloning system and method fusing rhythm characteristics, belongs to the technical field of voice synthesis and natural language, and is applied to the aspect of fine-grained rhythm control in zero-sample voice synthesis. The implementation method comprises the following steps of: 1, extracting rhythm features and audio features of an audio file, and further respectively acquiring pause features, speed features and tone features in the rhythm features by sequentially adopting transcriptional text inverse coding, syllable-level speed registration quantization and pitch sequence feature splicing modes; 2, fusing the features of the audio files in a manner of discarding feature screening without guidance of a classifier; 3, generating a target Mel spectrogram based on conditional flow matching; generating a target audio file from a to-be-cloned audio file through the trained voice cloning model controlled by the fusion rhythm; compared with the prior art, fine-grained rhythm control of tone and rhythm feature decoupling is realized in zero-sample speech synthesis, so that intonation accuracy based on a context scene is improved.

The present disclosure provides a pitch-based speech conversion model training method and a speech conversion system, wherein an audio feature code is output by a priori encoder, and a pitch feature is extracted by a pitch extraction module. A linear spectrum corresponding to the reference speech is input into the posteriori encoder to obtain an audio latent variable. In addition, the audio feature code, a speech concatenation feature obtained by concatenation of the audio feature code and the pitch feature, and the audio latent variable are input into a temporal alignment module to obtain a converted speech code, and the converted speech code is decoded by a decoder to obtain a converted speech. The training loss of the converted speech is then calculated to determine the degree of convergence of the speech conversion model.

A system for creating a song using vocal attributes extracted from a singing voice of an individual is provided comprising a computer and application executing thereon that digitally captures a plurality of vocal attributes of a singing voice of an individual. The system also processes, normalizes, identifies, and stores the attributes as a voice model. Based on dynamic outputs and digital audio files associated with the model, the system generates a performance of a new song created with the individual's singing voice. The system provides an identifier code for the voice model for storing and indexing the voice model and metadata or tags for identifying and tracking use of the voice model. The vocal attributes comprise at least range, timbre, flexibility, control, vibrato, resonance, articulation, expressiveness, stamina, tone and power of the singing voice and vocal traits comprising use of at least one of feel, phrasing, pronunciations, language, articulation, dynamics, meter, and rhythm. The attributes are captured from a live session with the individual or a recording of the individual's voice.

The invention relates to the technical field of artificial intelligence and the field of financial science and technology, and discloses a multi-module collaborative voice generation method, which comprises the following steps: acquiring a reference text and reference voice, converting the reference text into a phoneme sequence and generating potential features, converting the reference voice into a Mel spectrogram and extracting style features, and extracting fundamental frequency information. A phoneme sequence and a Mel spectrogram are input into a text alignment module to generate alignment information, voice duration is predicted based on the alignment information and potential features, fundamental frequency of voice is predicted through fundamental frequency information, the predicted duration, fundamental frequency and style features are input into a decoder to generate a reconstructed Mel spectrogram, and finally the reconstructed Mel spectrogram is converted into voice audio data through a vocoder. According to the method, the generated voice can better reflect the natural rhythm and tone change of the text content by combining accurate control on the voice duration, the fundamental frequency and the style characteristics, and the rhythm and expressive force of the voice are enhanced, so that the naturalness and the fidelity of voice synthesis are improved.

The present disclosure provides a system and a method for voice based authentication, which involves receiving voice data of a user enunciating a predetermined sequence of speech elements, with the predetermined sequence of speech elements is configured to capture a defined range of voiced sounds produced by the user; extracting voice features from the received voice data; deriving a voice signature for the user based on the extracted voice features, wherein the voice signature is representative of at least one of tonal, timbral, and temporal characteristics of the user's voiced speech; storing the derived voice signature in a database; receiving a verification voice sample of the user enunciating a predetermined sub-set of speech elements; comparing the verification voice sample with the stored voice signature; and authenticating the user based on the comparison.

A loudspeaker system and method can measure a specific user's hearing and implement compensatory or corrective processing to address the user's hearing deficiencies. The user's hearing acuity may be determined by (i) headphone techniques in which frequency-based hearing thresholds are determined; (ii) a loudspeaker system set up as intended for use in an acoustic space to emit tonal stimuli; or (iii) inducing, acquiring, and interpreting otoacoustic emissions. Once the hearing profile is determined, appropriate signal processing parameters can be used to generate a corrected audio output signal adjustment. The administered hearing test results can include a plurality of hearing loudness threshold levels at respective test frequencies. The hearing correction system's corrected audio output can be generated from corrected loudness values at selected correction frequency sub-bands.

The invention relates to the technical field of voice recognition, in particular to a voice interaction system and method for customer service based on artificial intelligence, and the system comprises a voice input processing module, an intention classification and routing module, a context dynamic adjustment module, a user behavior learning module, a multi-level intention fusion module and a final result module. According to the method, a multi-dimensional feature system is constructed by extracting tone intensity, speech speed frequency and emotional fluctuation amplitude, intention categories, priority weights and confidence scores are generated to realize accurate acquisition of appeals, and dialogue history, context and emotional change dynamic reconstruction path nodes, switching rules and response time sequences are tracked during interaction. Historical behavior mining preference features, habit fusion intention relevance, emergency calculation of an optimal strategy, construction of service steps, resource allocation schemes and execution timelines, adjustment of an interactive interface, a service process and a feedback mechanism according to multi-dimensional analysis, guarantee of differentiated service experience, and improvement of response accuracy and user satisfaction.

The invention belongs to the technical field of audio equipment design, and relates to a microphone for AI voice cloning based on pseudo tone of acoustic shielding, and the microphone comprises an audio collection module which is used for collecting an environment audio signal in real time; the analog-to-digital converter is used for converting the environment audio signal into a digital signal; the feature extraction module is used for performing frequency domain conversion on the digital signals and extracting frequency spectrum features and tone features; the acoustic shielding calculation module is used for calculating a shielding threshold value of the environment audio signal and determining a frequency band which cannot be heard by human ears in the environment audio signal according to the shielding threshold value; the pseudo tone generation module is used for generating pseudo tones based on the extracted tone features; the synthesis module is used for replacing a corresponding frequency band of the environment audio signal with the generated pseudo tone to form a synthesized audio; and the audio output module is used for outputting the synthesized audio. According to the invention, the pseudo tone can be generated in real time in the audio recording process, and it is ensured that a user is always protected during audio recording.

A musical instrument with novel configurations and performance techniques that can play notes and scale tones is disclosed. The musical instrument implements design and ergonomic changes to a Theremin's traditional configuration and use. The musical instrument includes an additional sound generation device such as an actuator that is operated to provide rhythm and tonal results. The musical instrument includes a pitch antenna that is configured in a horizontal position, which creates a different plane of reference with respect to performance motions employed for modifying the pitch. The musical instrument allows for a user to play notes and scale tones in a standing, seated, or ambulatory position. Improved playing interfaces, accessory items, and other characteristics for enhancing the playing of the musical instrument are provided.

The invention discloses a voice recognition emergency braking system and method for an intelligent driving vehicle, relates to the technical field of voice recognition emergency braking, and solves the problem that when emergency braking is performed through voice recognition, the situation that in-vehicle volumes of different users have different characteristics in different time periods is neglected, and emergency braking is performed according to the different characteristics, so that the emergency braking efficiency is improved. And the technical problem of difficulty in self-optimization of the time period in division is solved. The method comprises the following steps: acquiring sound data when a vehicle runs; dynamically planning the time period of each day, analyzing the historical volume in the vehicle to obtain the characteristic volume of each time period in the vehicle, and obtaining a discrimination factor based on the characteristic volume, the volume in the vehicle, the personnel tone and the sound data outside the vehicle; obtaining a control instruction according to the personnel voice characters and the discrimination factor; the vehicle is braked based on the control instruction; according to the invention, the self-adaptability and accuracy of the voice recognition emergency braking system of the intelligent driving vehicle can be improved.

The invention provides a method and system for determining reply verbal skills of an outbound robot, and relates to the technical field of conversation robotics.The method comprises the steps that tone, volume and voice activity rate features of user voice signals are extracted, and voice feature vectors are generated; converting the voice signal into a text, generating a text semantic vector, and combining to generate a user state vector; generating a context semantic vector and an intention semantic vector based on a RoBERTa language model, calculating an intention correlation score, calculating an intention completion rate through a prediction model, generating a priority score in combination with an emergency degree, selecting a user intention according to a priority score function, training a reply model based on historical verbal skill data, and obtaining a reply result. The most appropriate recovery verbal skill is generated in real time, and the intelligence and adaptability of robot recovery are improved.

The embodiment of the invention provides a voice optimization method and device, electronic equipment and a storage medium, and belongs to the technical field of voice optimization and artificial intelligence. The method comprises the following steps: acquiring current audio data of a target object, and performing feature extraction on the current audio data by using a pre-constructed target voice optimization model to obtain current timbre features of the target object; obtaining target music score data according to the current audio data, and performing feature extraction on the target music score data to obtain target tone features; performing feature splicing on the current tone feature and the target tone feature to obtain a spliced audio feature; performing random noise adding processing on the spliced audio features to obtain noise-added audio features; and performing voice optimization on the current audio data by using the target voice optimization model based on the noise-added audio characteristics. According to the embodiment of the invention, the voice optimization efficiency can be improved.

The invention relates to the technical field of voice explanation, in particular to an intelligent interactive voice explanation system based on an AIGC large model, which comprises an intelligent interactive voice explanation device main body 1, a memory and a processor are carried in the intelligent interactive voice explanation device main body 1, and a computer program is stored in the memory. The processor executes a computer program to realize the intelligent interactive voice explanation system based on the AIGC large model. The system comprises a tourist intention recognition module, a standard commentary matching module, an explanation content supplementing module and an explanation content broadcasting module. According to the invention, through converting the voice of the tourist into the text and analyzing the demand of the tourist, the matched standard explanation content can be identified in time, the correlation and pertinence of the explanation content are improved by analyzing the information coverage rate of the explanation and searching for the supplementary content, and the naturalness of voice output is enhanced by adjusting the speech speed and tone of the voice, so that the user experience is improved. And the interactive experience of tourists and the convenience of information acquisition are enriched.

An electronic instrument includes a casing, a controller having multiple operation elements provided in connection with the casing and aligned in a left-right direction of the casing, a sound emitting unit including a low-pitched sound emitting unit provided on a first side in the left-right direction in the casing and configured to emit a low-pitched sound and a high-pitched sound emitting unit provided on a second side in the left-right direction in the casing and configured to emit a high-pitched sound, and a reflection unit having a reflection surface configured to reflect a sound and provided between the low-pitched sound emitting unit and the high-pitched sound emitting unit in the casing, and the reflection unit is configured such that the reflection surface is inclined towards the high-pitched sound emitting unit with respect to the left-right direction.

The invention provides a music generation method, a music generation device, electronic equipment and a storage medium, and the music generation method comprises the steps: after receiving a music generation instruction, generating an accompaniment audio and a singing audio according to the music generation instruction, carrying out the audio synthesis of the accompaniment audio and the singing audio, and generating and outputting target music. Therefore, when the method is applied to the vehicle intelligent cabin scene, the accompaniment audio and the singing audio are respectively generated by adopting an artificial intelligence technology, and the accompaniment audio and the singing audio are synthesized to obtain the target music, so that the adaptation degree of the accompaniment audio and the singing audio in music content creation can be effectively improved; the accompaniment audio and the singing audio are closely fused in multiple dimensions such as rhythm, tone and emotional expression, so that the user experience is greatly improved.

The invention relates to the field of intelligent voice recognition, and particularly discloses a voice style recognition system based on a large model, which is characterized in that a voice input signal of a target user acquired by a microphone is received, and then a Mel spectrum is extracted from the voice input signal; a signal analysis and processing technology based on deep learning and a large model is adopted to carry out block processing and multi-scale semantic coding on the extracted voice input signal Mel spectrogram; therefore, the voice style recognition broadcast voice is intelligently generated according to the dynamic feature aggregation representation between the context Mel spectrum local coding features obtained through coding. Through the mode, the system can ensure that complex styles are fully represented, and can better adapt to significant differences of different speakers in the aspects of pronunciation modes, speech speeds, tones and the like, and the generalization ability is improved.

The invention discloses a voice interaction method, a server and a computer readable storage medium. The method comprises: determining an emotional state identifier according to acoustic features of a received voice request, the acoustic features including at least one of a tone feature, a speech speed feature and an energy feature; and according to the emotional state identifier and the voice request, generating an emotional adaptive pad call so as to complete the voice interaction. Therefore, by analyzing the acoustic features, identifying the emotional state identifier, accurately generating the emotional adaptive pad call corresponding to the voice request, and establishing emotional connection with the user, the user experience is enhanced. Moreover, by generating the emotion-adaptive pad call, a waiting blank period from the time after the user sends an instruction to the time before the system returns a formal result can be filled up, the experience of'instant response 'is created, the user is prevented from anxiety caused by too long waiting time, and the satisfaction degree and the credibility of the system are further improved.

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 provides a multi-mode dysarthria speech reconstruction system and a multi-mode dysarthria speech reconstruction method. The system comprises an encoder and a decoder. The encoder is configured to: extract a multi-modal feature from a multi-modal input of dysarthria speech and generate phoneme embedding based on the multi-modal feature; predicting a sequence of phoneme durations based on the phoneme embedding and generating an extended phoneme embedding and a predicted tone sequence; and extracting a speaker identity characterization of the dysarthria speech speaker. The decoder is configured to reconstruct a normal speech waveform corresponding to dysarthria speech with the extended phoneme embedding, the predicted tone sequence, and the speaker identity representation as inputs.

The invention discloses a personalized voice content generation method, which comprises the following steps: acquiring a voice text to be generated, and generating a preliminary tone distribution diagram; according to a preset pause length parameter, generating a voice rhythm model including pause duration; integrating the tone distribution diagram with the voice rhythm model, and judging whether an area where tone change conflicts with pause length exists or not; extracting mood word distribution information in the text, and dynamically adjusting the use frequency of mood words by combining accent position parameters and analyzing preferences and use scenes of the user; comprehensively considering the mutual influence of tone change, pause length, accent position and mood frequency, and generating a dynamically adjusted voice parameter model; combining the dynamically adjusted voice parameter model with a text, and extracting tone, pause and accent features in a waveform; judging a natural degree index in the voice waveform data, and optimizing the stability of voice output; noise and unstable fluctuation are eliminated, and final personalized voice output is generated.

The invention belongs to the technical field of emotion music generation, and particularly relates to a cross-modal music automatic generation system and method based on emotion recognition, and the method comprises the steps: synchronously collecting the facial expression, voice tone and ECG physiological signals of a user through a signal collection unit; processing the collected information through a multi-modal emotion recognition model to obtain a VAD three-dimensional continuous emotion vector, inputting the VAD three-dimensional continuous emotion vector into a music generation module, and constructing a shared cross-modal potential space through an emotion auto-encoder and a music auto-encoder in the music generation module; constraining the consistency of emotion-music in the potential space by adopting a comparative learning loss function; and generating a music file in an MIDI format based on the Mus-Decoder. The system can fully combine facial expressions, voice tones and ECG physiological signal multi-mode modes to generate music matched with the current emotion of the user, and emotion semantic consistency is achieved.

To provide a speech recognition system with higher inference accuracy by using a large-scale language model.SOLUTION: A speech recognition system includes: a first pre-trained inference model (SSL model 20) that can output tokens corresponding to words contained in speech signals on a token-by-token basis, a second pre-trained inference model (large-scale language model 30) that outputs one or more corresponding tokens on the basis of inference based on one or more token inputs, and a multiplier 26 that multiplies a first inference output 22 of the first pre-trained inference model, and a second inference output 32 corresponding to the first inference output, which is an output of the second pre-trained inference model, on a word-by-word basis. Then, the word selected based on the new probability distribution, which is an output of the multiplier, is input to the second pre-trained inference model.SELECTED DRAWING: Figure 3

The invention discloses an AI-based emotional text speech conversion method and device. The method comprises the following steps: acquiring speech segments and text records from historical data of a user; performing noise reduction processing and feature extraction according to the voice segments and the text records to obtain voice features; inputting the voice features into a pre-constructed emotional tendency model, and outputting emotional tendency and emotional intensity; according to the emotional tendency and the emotional intensity, adjusting a tone weight, a speech speed and a volume to obtain a speech parameter; extracting new voice features according to the voice parameters to perform scene emotion label matching, and determining voice adjustment parameters through a linear regression model; according to the emotional tendency and the new voice features, generating an emotional type through a pre-established emotional intention classification model, and calculating a voice parameter weight in combination with a pre-established emotional mapping table; and according to the voice adjustment parameter and the voice parameter weight, performing language synthesis to generate personalized voice. According to the method, personalized expression can be accurately generated according to the scene.

The invention discloses a wireless direction finder circuit and a wireless direction finder, and relates to the technical field of wireless direction finding. Wherein the wireless direction finder circuit comprises a local oscillation signal input end, a frequency mixing detection circuit, a rectifying circuit, a main control circuit and an inflexion output circuit. The frequency mixing detection circuit generates a frequency mixing signal according to the wireless signal and the local oscillator signal. The rectifying circuit converts the frequency mixing signal into a first direct current signal. And the main control circuit outputs an inflexion control signal when the voltage value of the first direct current signal is greater than or equal to a preset voltage value. When the inflexion output circuit does not receive the inflexion control signal, the inflexion output circuit outputs a corresponding first audio signal according to the frequency mixing signal; and when the inflexion control signal is received, the second audio signal of frequency conversion is output according to the inflexion control signal and the frequency mixing signal, so that a user can be prompted that the signal intensity is relatively high at the moment and the signal source has entered the preset distance range. According to the invention, positioning is assisted through a dual change mechanism of tone and volume, so that positioning efficiency and environmental adaptability are improved.

Devices, systems, and methods are disclosed for customizing audio experiences associated with operation of a ridable device. An electronic device having a processor detects contextual information such as speed, acceleration, ambient noise level, environment type, or location of the ridable device. Sensor data from accelerometers, gyroscopes, microphones, cameras, or positioning systems may be processed to identify these conditions. Based on the detected context, the electronic device determines audio configuration parameters, including selection of audio content, adjustment of volume, blending of tracks, or modification of tempo or pitch. The determined parameters are applied to output audio through one or more speakers of the electronic device, the ridable device, or associated accessories. In some implementations, playback is synchronized with other riders using a common time reference, geo-fence triggers are employed to provide location-specific content, or warning sounds are generated when nearby objects are detected, enhancing both rider enjoyment and environmental awareness.