A voicing decision algorithm using continuous nonlinear network is discussed. A five dimensional feature vector is used to describe the voicing characteristic of speech segment, and a continuous network is trained wi...A voicing decision algorithm using continuous nonlinear network is discussed. A five dimensional feature vector is used to describe the voicing characteristic of speech segment, and a continuous network is trained with a gradient descent algorithm is served as the voicing decision maker. Computer simulation shows that this algorithm is an outperform way to make voicing decision. The correct rate of this method reaches 97.8%.展开更多
“We are together.We are friends forever.Nothing can break the bond between you and me,”On December 30,2025,these heartfelt lyrics,sung in both Chinese and English,filled the historic Erqi Theater in central Beijing....“We are together.We are friends forever.Nothing can break the bond between you and me,”On December 30,2025,these heartfelt lyrics,sung in both Chinese and English,filled the historic Erqi Theater in central Beijing.Theywere performed by 54 young vocalists from the One Voice Children’s Choir of Utah,U.S.,joined on stage by a dozen students from the High School Affiliated to Xi’an Jiaotong University.Ranging in age from five to 18,the choir delivered a spectacular performance that visibly captivated the audience.展开更多
With the popularization of new technologies,telephone fraud has become the main means of stealing money and personal identity information.Taking inspiration from the website authentication mechanism,we propose an end-...With the popularization of new technologies,telephone fraud has become the main means of stealing money and personal identity information.Taking inspiration from the website authentication mechanism,we propose an end-to-end datamodem scheme that transmits the caller’s digital certificates through a voice channel for the recipient to verify the caller’s identity.Encoding useful information through voice channels is very difficult without the assistance of telecommunications providers.For example,speech activity detection may quickly classify encoded signals as nonspeech signals and reject input waveforms.To address this issue,we propose a novel modulation method based on linear frequency modulation that encodes 3 bits per symbol by varying its frequency,shape,and phase,alongside a lightweightMobileNetV3-Small-based demodulator for efficient and accurate signal decoding on resource-constrained devices.This method leverages the unique characteristics of linear frequency modulation signals,making them more easily transmitted and decoded in speech channels.To ensure reliable data delivery over unstable voice links,we further introduce a robust framing scheme with delimiter-based synchronization,a sample-level position remedying algorithm,and a feedback-driven retransmission mechanism.We have validated the feasibility and performance of our system through expanded real-world evaluations,demonstrating that it outperforms existing advanced methods in terms of robustness and data transfer rate.This technology establishes the foundational infrastructure for reliable certificate delivery over voice channels,which is crucial for achieving strong caller authentication and preventing telephone fraud at its root cause.展开更多
Parkinson’s disease remains a major clinical issue in terms of early detection,especially during its prodromal stage when symptoms are not evident or not distinct.To address this problem,we proposed a new deep learni...Parkinson’s disease remains a major clinical issue in terms of early detection,especially during its prodromal stage when symptoms are not evident or not distinct.To address this problem,we proposed a new deep learning 2-based approach for detecting Parkinson’s disease before any of the overt symptoms develop during their prodromal stage.We used 5 publicly accessible datasets,including UCI Parkinson’s Voice,Spiral Drawings,PaHaW,NewHandPD,and PPMI,and implemented a dual stream CNN–BiLSTM architecture with Fisher-weighted feature merging and SHAP-based explanation.The findings reveal that the model’s performance was superior and achieved 98.2%,a F1-score of 0.981,and AUC of 0.991 on the UCI Voice dataset.The model’s performance on the remaining datasets was also comparable,with up to a 2–7 percent betterment in accuracy compared to existing strong models such as CNN–RNN–MLP,ILN–GNet,and CASENet.Across the evidence,the findings back the diagnostic promise of micro-tremor assessment and demonstrate that combining temporal and spatial features with a scatter-based segment for a multi-modal approach can be an effective and scalable platform for an“early,”interpretable PD screening system.展开更多
Background:This study evaluates the ability of mobile AI voice assistants(AI-VAs)to provide accurate medical advice for early knee osteoarthritis(KOA)and compares their performance with conventional web searches and h...Background:This study evaluates the ability of mobile AI voice assistants(AI-VAs)to provide accurate medical advice for early knee osteoarthritis(KOA)and compares their performance with conventional web searches and human clinicians.Methods:From September to October 2024,two AI-VAs(Apple’s Siri and Huawei’s Xiaoyi)were tested on 15 KOA-related questions in Chinese and English.The assessment focused on the accuracy of voice recognition,response capabilities,and medical advice.Siri was further tested in four international regions(USA,UK,Germany,Hong Kong)using localized languages.Results:In Chinese-language tests,Siri and Xiaoyi showed comparable voice recognition(recognition accuracy:95.6%vs.93.3%)and response ability(speech response:88.9%vs.85.7%).However,Siri provided significantly more accurate medical advice(medical advice:95.6%vs.53.3%;Z=2.762,P<0.001).External validation via Global Quality Score further confirmed Siri’s superiority(mean Global Quality Score=4.0 vs.Xiaoai=0.9).Siri outperformed Xiaoyi in English-language tests(53.3%vs.0%).While Siri’s medical advice accuracy(95.6%)surpassed non-specialist clinicians(Z=2.685,P=0.007),it primarily reflects filtered search results(Baidu/Google)rather than clinical synthesis.Claims of equivalence to junior surgeons(98.2%)must be interpreted cautiously,as AI-VAs lack diagnostic reasoning capabilities.This distinction is critical to avoid overstating their role in clinical decision-making.Conclusion:Current AI-VAs offer limited value in providing precise medical advice for KOA,primarily serving as intermediaries for web search results.Their performance varies across languages,regions,and search engines.展开更多
Non-Terrestrial Networks(NTN)can be used to provide emergency voice services in Sixth-Generation(6G)communication systems.However,Internet of Things(Io T)terminals,which comprise restricted bandwidth resources and wea...Non-Terrestrial Networks(NTN)can be used to provide emergency voice services in Sixth-Generation(6G)communication systems.However,Internet of Things(Io T)terminals,which comprise restricted bandwidth resources and weak computing power,which make ensuring high-quality voice services over NTN challenging.Recent advancements in Artificial Intelligence(AI)techniques have been increasingly applied to enhance the audio quality and reduce the bit rate.However,applying models with high computational complexity to Io T terminals is difficult.In this study,we propose a voice-services-over NTN solution including a novel 6G non-terrestrial and ground network integrated framework and a lightweight Large Models(LMs)-driven codec operating at 450 bits per second.We also designed a new voice packet header and deployed an agent on-ground gateway to reduce the bandwidth overhead.The non-standard Session Initiation Protocol header was converted to the standard format while re-encapsulating Internet Protocol and User Datagram Protocol headers,replacing the conventional implementations.Additionally,an operational NTN satellite was used to evaluate the proposed Re Codec.The experimental results demonstrate that the Re Codec decreases the computational complexity by 96.61%while increasing the voice quality by 17.55%when compared with the state-of-the-art mechanisms.Furthermore,the design of the packet header reduced the voice frame header to 50 bytes.展开更多
Du Zhanyuan,Standing Committee member of the 14th CPPCC National Committee and CICG president,on how to tell engaging stories about China.AS changes unseen in a century accelerate across the world,cultural exchange an...Du Zhanyuan,Standing Committee member of the 14th CPPCC National Committee and CICG president,on how to tell engaging stories about China.AS changes unseen in a century accelerate across the world,cultural exchange and mutual learning between civilizations are becoming increasingly important.展开更多
The kingdom Animalia encompasses multicellular,eukaryotic organisms known as animals.Currently,there are approximately 1.5 million identified species of living animals,including over 195 distinct breeds of dogs.Each b...The kingdom Animalia encompasses multicellular,eukaryotic organisms known as animals.Currently,there are approximately 1.5 million identified species of living animals,including over 195 distinct breeds of dogs.Each breed possesses unique characteristics that can be challenging to distinguish.Each breed has its own characteristics that are difficult to identify.Various computer-based methods,including machine learning,deep learning,transfer learning,and robotics,are employed to identify dog breeds,focusing mainly on image or voice data.Voice-based techniques often face challenges such as noise,distortion,and changes in frequency or pitch,which can impair the model’s performance.Conversely,image-based methods may fail when dealing with blurred images,which can result from poor camera quality or photos taken from a distance.This research presents a hybrid model combining voice and image data for dog breed identification.The proposed method Contrastive Language-Image Pre-Training-Audio Stacked Network(CLIP-ASN)improves robustness,compensating when one data type is compromised by noise or poor quality.By integrating diverse data types,the model can more effectively identify unique breed characteristics,making it superior to methods relying on a single data type.The key steps of the proposed model are data collection,feature extraction based on Contrastive Language Image for image-based feature extraction and Audio stacked-based voice features extraction,co-attention-based classification,and federated learning-based training and distribution.From the experimental evaluation,it has been concluded that the performance of the proposed work in terms of accuracy 89.75%and is far better than the existing benchmark methods.展开更多
This study focuses on the elderly memo app“Remembering,”addressing memory decline and operational difficulties.It introduces a progressive interaction system with three core modules:Dynamic font adjustment,intellige...This study focuses on the elderly memo app“Remembering,”addressing memory decline and operational difficulties.It introduces a progressive interaction system with three core modules:Dynamic font adjustment,intelligent voice reminders,and family warning systems.Health monitoring and remote care functions are also integrated,creating a simple operation process.The research highlights four design dimensions for elderly-friendly products:Usability,security,emotionalization,and personalization.This innovation reduces the digital barrier and provides a model for smart elderly-friendly product development.展开更多
This study examined the relationship between inclusive leadership and subordinates’upward voice,focusing on the mediating role of psychological safety and the moderating role of collectivism.Data were collected from ...This study examined the relationship between inclusive leadership and subordinates’upward voice,focusing on the mediating role of psychological safety and the moderating role of collectivism.Data were collected from 284 subordinates and supervisors across 11 organizations in China in three cross-lagged waves.Structural equation modeling results indicated that inclusive leadership was associated with subordinates’upward voice via psychological safety.Moreover,collectivism strengthens the association between inclusive leadership and upward voice via psychological safety,leading to a higher upward voice.These findings highlight the importance of inclusive leadership in fostering an environment that promotes open communication and psychological safety between supervisors and subordinates,ultimately enhancing workplace health and well-being.The implications of these findings suggest that management practices should cultivate inclusive leadership behaviors for enhancing psychological safety,and encouraging subordinates to voice their opinions for the overall success of the organization.展开更多
In recent years,Speech Emotion Recognition(SER)has developed into an essential instrument for interpreting human emotions from auditory data.The proposed research focuses on the development of a SER system employing d...In recent years,Speech Emotion Recognition(SER)has developed into an essential instrument for interpreting human emotions from auditory data.The proposed research focuses on the development of a SER system employing deep learning and multiple datasets containing samples of emotive speech.The primary objective of this research endeavor is to investigate the utilization of Convolutional Neural Networks(CNNs)in the process of sound feature extraction.Stretching,pitch manipulation,and noise injection are a few of the techniques utilized in this study to improve the data quality.Feature extraction methods including Zero Crossing Rate,Chroma_stft,Mel⁃scale Frequency Cepstral Coefficients(MFCC),Root Mean Square(RMS),and Mel⁃Spectogram are used to train a model.By using these techniques,audio signals can be transformed into recognized features that can be utilized to train the model.Ultimately,the study produces a thorough evaluation of the models performance.When this method was applied,the model achieved an impressive accuracy of 94.57%on the test dataset.The proposed work was also validated on the EMO⁃BD and IEMOCAP datasets.These consist of further data augmentation,feature engineering,and hyperparameter optimization.By following these development paths,SER systems will be able to be implemented in real⁃world scenarios with greater accuracy and resilience.展开更多
During the 2025 Beijing Digital Economy Experience Week from 27 June to 5 July,interactive projects combining technology and culture-illustrated books created by voice AI,augmented reality tours,and markerless motion ...During the 2025 Beijing Digital Economy Experience Week from 27 June to 5 July,interactive projects combining technology and culture-illustrated books created by voice AI,augmented reality tours,and markerless motion capture-attracted many visitors.“More than a dozen themed areas offered the opportunity to dive into new worlds and discover the latest innovations from more than 50 companies,”Lu Yumin,a Beijing resident who visited the event,told ChinAfrica.展开更多
Objectives:By investigating the distinct speech and voice phenotype among TCM constitution for adults,this study aims at providing a convenient and objective methodological reference for judging TCM constitution.Metho...Objectives:By investigating the distinct speech and voice phenotype among TCM constitution for adults,this study aims at providing a convenient and objective methodological reference for judging TCM constitution.Methods:Acoustic analysis and TCM constitution assessment were performed for all 620 participants using Praat software and the CCMQ,respectively.Results:For formant features,the speech duration of special constitution participants was shorter than that of neutral,phlegm-dampness,dampness-heat,Yin-deficiency,or Yang-deficiency participants when pronuncing the vowels/a/,/i/,and/u/.Compare to Yang-deficiency,Qi-deficiency participants had a shorter speech duration when pronucing/i/.For/u/,blood-stasis participants exhibited a lower F1 value than neutral participants.For vocal features,special constitution participants showed higher local jitter than neutral,dampness-heat,and Yang-deficiency participants(for/a/,/i/,and/u/).Higher absolute local jitter than neutral or dampness-heat participants.Compared with neutral or Yang-deficiency participants,special participants owned a higher local shimmer(dB).Special participants had a lower harmonicity autocorrelation than neutral,dampness-heat,or Yang-deficiency participants.Conclusions:Formant features may effectively differentiate special constitution from neutral,phlegm-dampness,dampness-heat,Yin-deficiency,or Yang-deficiency constitutions based on vowel duration measurements(/a/,/i/,/u/).For the vowel/u/,F1 values may help distinguish blood-stasis from neutral constitution.Vocal features appear particularly useful for distinguishing special constitution from neutral,dampness-heat,or Yang-deficiency constitution,with local jitter and harmonicity autocorrelation showing significant discriminatory power.展开更多
This study examined the relationship between leader-employee calling congruence on employees’voice behaviour.Participants were 173 leader-employee dyads from the Chinese service industry.They completed online surveys...This study examined the relationship between leader-employee calling congruence on employees’voice behaviour.Participants were 173 leader-employee dyads from the Chinese service industry.They completed online surveys on calling,perceived insider status,and voice behaviour.Results from polynomial regression and response surface analysis showed that employees perceived insider status to be weaker with the low leader-low subordinate calling congruence,and stronger with high leader and high subordinate calling congruence.Employees perceived insider status is stronger in low leader and high subordinate calling incongruence compared with high leader and low subordinate calling incongruence.Perceived insider status plays a mediating role among calling congruence and voice behaviour.This study’sfindings suggest pathways of calling congruence on voice behaviour,which are important for promoting employee voice behaviour and guiding organisational recruitment in the workplace.展开更多
Mobile robots represented by smart wheelchairs can assist elderly people with mobility difficulties.This paper proposes a multi-mode semi-autonomous navigation system based on a local semantic map for mobile robots,wh...Mobile robots represented by smart wheelchairs can assist elderly people with mobility difficulties.This paper proposes a multi-mode semi-autonomous navigation system based on a local semantic map for mobile robots,which can assist users to implement accurate navigation(e.g.,docking)in the environment without prior maps.In order to overcome the problem of repeated oscillations during the docking of traditional local path planning algorithms,this paper adopts a mode-switching method and uses feedback control to perform docking when approaching semantic goals.At last,comparative experiments were carried out in the real environment.Results show that our method is superior in terms of safety,comfort and docking accuracy.展开更多
Objective:The objective of this study was to compare the effect of nurse and beloved family member’s recording voice on consciousness and physical parameters in patients with coma state.Materials and Methods:A random...Objective:The objective of this study was to compare the effect of nurse and beloved family member’s recording voice on consciousness and physical parameters in patients with coma state.Materials and Methods:A randomized control trial parallel group design was conducted among 45 comatose patients divided into two intervention groups,i.e.nurse voice stimulus group,receiving nurses voice with standard care,family members voice stimulus group receiving their beloved family member voice with standard care and one control group receiving only standard care in medicine intensive care unit.The intervention was provided three times a day,each lasting 5 min for 7 days in addition to standard care.Repeated measure analysis of variance and independent t-test were used to compare within and between groups,respectively.Results:The study found significant differences in Glasgow coma scale(GCS)scores within both the nurse(F=2.78,P=0.042)and family member(F=10.27,P=0.0001)voice stimulus groups over 7 days.Comparing GCS scores between intervention groups showed significant variations before(P=0.028),during(P=0.047),and after(P=0.036)the intervention on day 7.Comparing GCS scores between the family members’voice stimulus group and the control group,significant changes were observed on days 5 and 7(P=0.043,0.030,0.030,and 0.014,0.012,0.012)before,during,and after the intervention.Conclusions:The use of beloved family members’voices proved more effective in elevating the patients’level of consciousness compared to both the nurse voice stimulus group and the control group.展开更多
Although substantial research shows the effectiveness of written corrective feedback(WCF)in treating simple grammar structures,more research is still needed to refute Truscott’s claim that WCF may not work on complex...Although substantial research shows the effectiveness of written corrective feedback(WCF)in treating simple grammar structures,more research is still needed to refute Truscott’s claim that WCF may not work on complex grammar structures.Similarly,a previous body of research has shown that the degree of explicitness of feedback moderates the efficacy of WCF.However,most WCF studies have systematically manipulated only direct corrective feedback.The current study was therefore conducted to fill these gaps in the literature.To this end,five intact classes of Functional English were recruited and later randomly assigned to four treatment groups:DCF,DCF+ME,ICF,and ICF+ME,and one control group that received no feedback.All the groups took part in three WCF treatment sessions,during which they wrote two different pieces:a news report and a picture description.Later,only the treatment groups received the WCF.The WCF’s effectiveness was measured by writing tests and grammaticality judgment tasks(GJT).The results demonstrated that WCF helped L2 learners improve their grammatical accuracy of passive voice tenses.The study further showed that the group that received the most explicit type of WCF fared better than the ones that received the least explicit type of WCF.Important pedagogical implications for ESL/EFL teachers are discussed.展开更多
From children to teenagers and then to adults,individuals’emotional expression ability has undergone significant changes.From the compulsory education stage to the general senior middle school stage,students’cogniti...From children to teenagers and then to adults,individuals’emotional expression ability has undergone significant changes.From the compulsory education stage to the general senior middle school stage,students’cognitive and emotional needs will change greatly.Students of different school stages show significant differences in their vocal emotional expression,which not only affects their social skills but also has important guiding significance for educational practice.Therefore,this study focuses on the vocal emotional expression of students at different school stages.This study explores the differences in vocal emotional expression among students at different school stages(primary school,junior high school,and senior high school),with a focus on the expression characteristics of different emotional types and the situation of inconsistent internal and external emotions.The research results show that in terms of emotional types,students at different school stages have a relatively high accuracy rate in expressing sadness,while the accuracy rate for expressing anger is the lowest,and the accuracy rate for expressing anger is significantly lower than that for other emotions.In terms of school stages,the overall accuracy of emotional expression among students improves with the increase of school stages,but the differences between school stages are not significant.In the task of expressing inconsistent internal and external emotions,the study found that senior high school students have significantly better emotional conversion ability than primary and junior high school students,indicating that as the school stage increases,students’adaptability in complex emotional expression tasks gradually improves.展开更多
Background Vibrotactile feedback systems are widely used in assistive technology,wearable devices,and virtual environments to deliver precise tactile information.The timing of interstimulus intervals(ISIs)plays a crit...Background Vibrotactile feedback systems are widely used in assistive technology,wearable devices,and virtual environments to deliver precise tactile information.The timing of interstimulus intervals(ISIs)plays a critical role in determining how accurately users perceive and interpret vibrotactile patterns.The optimal use of ISIs can increase the effectiveness of these systems,improve user interaction,and enable reliable,intuitive feedback in diverse applications.We examined how different interstimulus intervals ISIs impact the accuracy of vibrotactile pattern recognition.Methods Participants wore a forearm-mounted device with six voice coil actuators arranged in a 3×2 grid,delivering Braille-based vibrotactile patterns sequentially at ISIs ranging from 10 to 2500 ms.Eight participants performed identification tasks involving Icelandic Braille patterns categorized as either short(2-3 actuators)or long(4-5 actuators).A repeated measures ANOVA was conducted to assess the effects of ISI,pattern type,and practice(across two testing blocks)on pattern recognition accuracy.Results For short patterns,accuracy was highest(92%-98%)at ISIs of 50-700 ms,with peak performance at 300 ms.For long patterns,accuracy reached 86%-94%at ISIs of 100-500 ms,peaking at 400 ms.Participants were more accurate with short patterns,and performance improved significantly over time for both short and long patterns,highlighting the importance of training for vibrotactile pattern recognition.Conclusions These results underscore the importance of careful selection of ISIs in vibrotactile feedback systems for accurate pattern identification.The findings provide valuable insights for conveying tactile information using wearable devices,contributing to better tactile feedback and performance in applications requiring precise vibrotactile information delivery.展开更多
This study introduces a novel voice cloning framework driven by Mordukhovich Subdifferential Optimization(MSO)to address the complex multi-objective challenges of pathological speech synthesis in underresourced Lithua...This study introduces a novel voice cloning framework driven by Mordukhovich Subdifferential Optimization(MSO)to address the complex multi-objective challenges of pathological speech synthesis in underresourced Lithuanian language with unique phonemes not present in most pre-trained models.Unlike existing voice synthesis models that often optimize for a single objective or are restricted to major languages,our approach explicitly balances four competing criteria:speech naturalness,speaker similarity,computational efficiency,and adaptability to pathological voice patterns.We evaluate four model configurations combining Lithuanian and English encoders,synthesizers,and vocoders.The hybrid model(English encoder,Lithuanian synthesizer,English vocoder),optimized via MSO,achieved the highest Mean Opinion Score(MOS)of 4.3 and demonstrated superior intelligibility and speaker fidelity.The results confirm that MSO enables effective navigation of trade-offs in multilingual pathological voice cloning,offering a scalable path toward high-quality voice restoration in clinical speech applications.This work represents the first integration of Mordukhovich optimization into pathological TTS,setting a new benchmark for speech synthesis under clinical and linguistic constraints.展开更多
文摘A voicing decision algorithm using continuous nonlinear network is discussed. A five dimensional feature vector is used to describe the voicing characteristic of speech segment, and a continuous network is trained with a gradient descent algorithm is served as the voicing decision maker. Computer simulation shows that this algorithm is an outperform way to make voicing decision. The correct rate of this method reaches 97.8%.
文摘“We are together.We are friends forever.Nothing can break the bond between you and me,”On December 30,2025,these heartfelt lyrics,sung in both Chinese and English,filled the historic Erqi Theater in central Beijing.Theywere performed by 54 young vocalists from the One Voice Children’s Choir of Utah,U.S.,joined on stage by a dozen students from the High School Affiliated to Xi’an Jiaotong University.Ranging in age from five to 18,the choir delivered a spectacular performance that visibly captivated the audience.
文摘With the popularization of new technologies,telephone fraud has become the main means of stealing money and personal identity information.Taking inspiration from the website authentication mechanism,we propose an end-to-end datamodem scheme that transmits the caller’s digital certificates through a voice channel for the recipient to verify the caller’s identity.Encoding useful information through voice channels is very difficult without the assistance of telecommunications providers.For example,speech activity detection may quickly classify encoded signals as nonspeech signals and reject input waveforms.To address this issue,we propose a novel modulation method based on linear frequency modulation that encodes 3 bits per symbol by varying its frequency,shape,and phase,alongside a lightweightMobileNetV3-Small-based demodulator for efficient and accurate signal decoding on resource-constrained devices.This method leverages the unique characteristics of linear frequency modulation signals,making them more easily transmitted and decoded in speech channels.To ensure reliable data delivery over unstable voice links,we further introduce a robust framing scheme with delimiter-based synchronization,a sample-level position remedying algorithm,and a feedback-driven retransmission mechanism.We have validated the feasibility and performance of our system through expanded real-world evaluations,demonstrating that it outperforms existing advanced methods in terms of robustness and data transfer rate.This technology establishes the foundational infrastructure for reliable certificate delivery over voice channels,which is crucial for achieving strong caller authentication and preventing telephone fraud at its root cause.
基金supported via funding from Prince Sattam bin Abdulaziz University project number(PSAU/2025/03/32440).
文摘Parkinson’s disease remains a major clinical issue in terms of early detection,especially during its prodromal stage when symptoms are not evident or not distinct.To address this problem,we proposed a new deep learning 2-based approach for detecting Parkinson’s disease before any of the overt symptoms develop during their prodromal stage.We used 5 publicly accessible datasets,including UCI Parkinson’s Voice,Spiral Drawings,PaHaW,NewHandPD,and PPMI,and implemented a dual stream CNN–BiLSTM architecture with Fisher-weighted feature merging and SHAP-based explanation.The findings reveal that the model’s performance was superior and achieved 98.2%,a F1-score of 0.981,and AUC of 0.991 on the UCI Voice dataset.The model’s performance on the remaining datasets was also comparable,with up to a 2–7 percent betterment in accuracy compared to existing strong models such as CNN–RNN–MLP,ILN–GNet,and CASENet.Across the evidence,the findings back the diagnostic promise of micro-tremor assessment and demonstrate that combining temporal and spatial features with a scatter-based segment for a multi-modal approach can be an effective and scalable platform for an“early,”interpretable PD screening system.
文摘Background:This study evaluates the ability of mobile AI voice assistants(AI-VAs)to provide accurate medical advice for early knee osteoarthritis(KOA)and compares their performance with conventional web searches and human clinicians.Methods:From September to October 2024,two AI-VAs(Apple’s Siri and Huawei’s Xiaoyi)were tested on 15 KOA-related questions in Chinese and English.The assessment focused on the accuracy of voice recognition,response capabilities,and medical advice.Siri was further tested in four international regions(USA,UK,Germany,Hong Kong)using localized languages.Results:In Chinese-language tests,Siri and Xiaoyi showed comparable voice recognition(recognition accuracy:95.6%vs.93.3%)and response ability(speech response:88.9%vs.85.7%).However,Siri provided significantly more accurate medical advice(medical advice:95.6%vs.53.3%;Z=2.762,P<0.001).External validation via Global Quality Score further confirmed Siri’s superiority(mean Global Quality Score=4.0 vs.Xiaoai=0.9).Siri outperformed Xiaoyi in English-language tests(53.3%vs.0%).While Siri’s medical advice accuracy(95.6%)surpassed non-specialist clinicians(Z=2.685,P=0.007),it primarily reflects filtered search results(Baidu/Google)rather than clinical synthesis.Claims of equivalence to junior surgeons(98.2%)must be interpreted cautiously,as AI-VAs lack diagnostic reasoning capabilities.This distinction is critical to avoid overstating their role in clinical decision-making.Conclusion:Current AI-VAs offer limited value in providing precise medical advice for KOA,primarily serving as intermediaries for web search results.Their performance varies across languages,regions,and search engines.
基金supported in part by the Major Key Project of PCL under Grant PCL2023A07the Research and Development Program of China Telecom under Grant T-2025-27。
文摘Non-Terrestrial Networks(NTN)can be used to provide emergency voice services in Sixth-Generation(6G)communication systems.However,Internet of Things(Io T)terminals,which comprise restricted bandwidth resources and weak computing power,which make ensuring high-quality voice services over NTN challenging.Recent advancements in Artificial Intelligence(AI)techniques have been increasingly applied to enhance the audio quality and reduce the bit rate.However,applying models with high computational complexity to Io T terminals is difficult.In this study,we propose a voice-services-over NTN solution including a novel 6G non-terrestrial and ground network integrated framework and a lightweight Large Models(LMs)-driven codec operating at 450 bits per second.We also designed a new voice packet header and deployed an agent on-ground gateway to reduce the bandwidth overhead.The non-standard Session Initiation Protocol header was converted to the standard format while re-encapsulating Internet Protocol and User Datagram Protocol headers,replacing the conventional implementations.Additionally,an operational NTN satellite was used to evaluate the proposed Re Codec.The experimental results demonstrate that the Re Codec decreases the computational complexity by 96.61%while increasing the voice quality by 17.55%when compared with the state-of-the-art mechanisms.Furthermore,the design of the packet header reduced the voice frame header to 50 bytes.
文摘Du Zhanyuan,Standing Committee member of the 14th CPPCC National Committee and CICG president,on how to tell engaging stories about China.AS changes unseen in a century accelerate across the world,cultural exchange and mutual learning between civilizations are becoming increasingly important.
基金King Saud University,Saudi Arabia for funding this work through Ongoing Research Funding Program(ORF-2025-953),King Saud University,Riyadh,Saudi Arabia.
文摘The kingdom Animalia encompasses multicellular,eukaryotic organisms known as animals.Currently,there are approximately 1.5 million identified species of living animals,including over 195 distinct breeds of dogs.Each breed possesses unique characteristics that can be challenging to distinguish.Each breed has its own characteristics that are difficult to identify.Various computer-based methods,including machine learning,deep learning,transfer learning,and robotics,are employed to identify dog breeds,focusing mainly on image or voice data.Voice-based techniques often face challenges such as noise,distortion,and changes in frequency or pitch,which can impair the model’s performance.Conversely,image-based methods may fail when dealing with blurred images,which can result from poor camera quality or photos taken from a distance.This research presents a hybrid model combining voice and image data for dog breed identification.The proposed method Contrastive Language-Image Pre-Training-Audio Stacked Network(CLIP-ASN)improves robustness,compensating when one data type is compromised by noise or poor quality.By integrating diverse data types,the model can more effectively identify unique breed characteristics,making it superior to methods relying on a single data type.The key steps of the proposed model are data collection,feature extraction based on Contrastive Language Image for image-based feature extraction and Audio stacked-based voice features extraction,co-attention-based classification,and federated learning-based training and distribution.From the experimental evaluation,it has been concluded that the performance of the proposed work in terms of accuracy 89.75%and is far better than the existing benchmark methods.
基金Liaoning Province Higher Education Institutions’College Students’Innovation and Entrepreneurship Training Program 2024 Class A Project,“Research on Intelligent Design from the Perspective of Elderly Adaptability-Taking the Elderly Memo as an Example”(X202413198121)。
文摘This study focuses on the elderly memo app“Remembering,”addressing memory decline and operational difficulties.It introduces a progressive interaction system with three core modules:Dynamic font adjustment,intelligent voice reminders,and family warning systems.Health monitoring and remote care functions are also integrated,creating a simple operation process.The research highlights four design dimensions for elderly-friendly products:Usability,security,emotionalization,and personalization.This innovation reduces the digital barrier and provides a model for smart elderly-friendly product development.
基金supported by the China Postdoctoral Science Foundation(Certificate Number:2024M760126).
文摘This study examined the relationship between inclusive leadership and subordinates’upward voice,focusing on the mediating role of psychological safety and the moderating role of collectivism.Data were collected from 284 subordinates and supervisors across 11 organizations in China in three cross-lagged waves.Structural equation modeling results indicated that inclusive leadership was associated with subordinates’upward voice via psychological safety.Moreover,collectivism strengthens the association between inclusive leadership and upward voice via psychological safety,leading to a higher upward voice.These findings highlight the importance of inclusive leadership in fostering an environment that promotes open communication and psychological safety between supervisors and subordinates,ultimately enhancing workplace health and well-being.The implications of these findings suggest that management practices should cultivate inclusive leadership behaviors for enhancing psychological safety,and encouraging subordinates to voice their opinions for the overall success of the organization.
文摘In recent years,Speech Emotion Recognition(SER)has developed into an essential instrument for interpreting human emotions from auditory data.The proposed research focuses on the development of a SER system employing deep learning and multiple datasets containing samples of emotive speech.The primary objective of this research endeavor is to investigate the utilization of Convolutional Neural Networks(CNNs)in the process of sound feature extraction.Stretching,pitch manipulation,and noise injection are a few of the techniques utilized in this study to improve the data quality.Feature extraction methods including Zero Crossing Rate,Chroma_stft,Mel⁃scale Frequency Cepstral Coefficients(MFCC),Root Mean Square(RMS),and Mel⁃Spectogram are used to train a model.By using these techniques,audio signals can be transformed into recognized features that can be utilized to train the model.Ultimately,the study produces a thorough evaluation of the models performance.When this method was applied,the model achieved an impressive accuracy of 94.57%on the test dataset.The proposed work was also validated on the EMO⁃BD and IEMOCAP datasets.These consist of further data augmentation,feature engineering,and hyperparameter optimization.By following these development paths,SER systems will be able to be implemented in real⁃world scenarios with greater accuracy and resilience.
文摘During the 2025 Beijing Digital Economy Experience Week from 27 June to 5 July,interactive projects combining technology and culture-illustrated books created by voice AI,augmented reality tours,and markerless motion capture-attracted many visitors.“More than a dozen themed areas offered the opportunity to dive into new worlds and discover the latest innovations from more than 50 companies,”Lu Yumin,a Beijing resident who visited the event,told ChinAfrica.
基金supported by the National Natural Science Foundation of China(Nos.81730107 and 81973883)the National Science&Technology Basic Research Project(No.2015FY111700)the Shanghai Pudong New District New Area Project(No.PW2022A-78(WQZ)).
文摘Objectives:By investigating the distinct speech and voice phenotype among TCM constitution for adults,this study aims at providing a convenient and objective methodological reference for judging TCM constitution.Methods:Acoustic analysis and TCM constitution assessment were performed for all 620 participants using Praat software and the CCMQ,respectively.Results:For formant features,the speech duration of special constitution participants was shorter than that of neutral,phlegm-dampness,dampness-heat,Yin-deficiency,or Yang-deficiency participants when pronuncing the vowels/a/,/i/,and/u/.Compare to Yang-deficiency,Qi-deficiency participants had a shorter speech duration when pronucing/i/.For/u/,blood-stasis participants exhibited a lower F1 value than neutral participants.For vocal features,special constitution participants showed higher local jitter than neutral,dampness-heat,and Yang-deficiency participants(for/a/,/i/,and/u/).Higher absolute local jitter than neutral or dampness-heat participants.Compared with neutral or Yang-deficiency participants,special participants owned a higher local shimmer(dB).Special participants had a lower harmonicity autocorrelation than neutral,dampness-heat,or Yang-deficiency participants.Conclusions:Formant features may effectively differentiate special constitution from neutral,phlegm-dampness,dampness-heat,Yin-deficiency,or Yang-deficiency constitutions based on vowel duration measurements(/a/,/i/,/u/).For the vowel/u/,F1 values may help distinguish blood-stasis from neutral constitution.Vocal features appear particularly useful for distinguishing special constitution from neutral,dampness-heat,or Yang-deficiency constitution,with local jitter and harmonicity autocorrelation showing significant discriminatory power.
基金supported by the Major Research Project of Philosophy and Social Sciences in Universities of Henan Province(2025-JCZD-10)the Natural Science Foundation of Henan Province(242300421311).
文摘This study examined the relationship between leader-employee calling congruence on employees’voice behaviour.Participants were 173 leader-employee dyads from the Chinese service industry.They completed online surveys on calling,perceived insider status,and voice behaviour.Results from polynomial regression and response surface analysis showed that employees perceived insider status to be weaker with the low leader-low subordinate calling congruence,and stronger with high leader and high subordinate calling congruence.Employees perceived insider status is stronger in low leader and high subordinate calling incongruence compared with high leader and low subordinate calling incongruence.Perceived insider status plays a mediating role among calling congruence and voice behaviour.This study’sfindings suggest pathways of calling congruence on voice behaviour,which are important for promoting employee voice behaviour and guiding organisational recruitment in the workplace.
基金the Technology Project Managed by the State Grid Corporation of China(No.5108-202218280A-2-249-XG)。
文摘Mobile robots represented by smart wheelchairs can assist elderly people with mobility difficulties.This paper proposes a multi-mode semi-autonomous navigation system based on a local semantic map for mobile robots,which can assist users to implement accurate navigation(e.g.,docking)in the environment without prior maps.In order to overcome the problem of repeated oscillations during the docking of traditional local path planning algorithms,this paper adopts a mode-switching method and uses feedback control to perform docking when approaching semantic goals.At last,comparative experiments were carried out in the real environment.Results show that our method is superior in terms of safety,comfort and docking accuracy.
文摘Objective:The objective of this study was to compare the effect of nurse and beloved family member’s recording voice on consciousness and physical parameters in patients with coma state.Materials and Methods:A randomized control trial parallel group design was conducted among 45 comatose patients divided into two intervention groups,i.e.nurse voice stimulus group,receiving nurses voice with standard care,family members voice stimulus group receiving their beloved family member voice with standard care and one control group receiving only standard care in medicine intensive care unit.The intervention was provided three times a day,each lasting 5 min for 7 days in addition to standard care.Repeated measure analysis of variance and independent t-test were used to compare within and between groups,respectively.Results:The study found significant differences in Glasgow coma scale(GCS)scores within both the nurse(F=2.78,P=0.042)and family member(F=10.27,P=0.0001)voice stimulus groups over 7 days.Comparing GCS scores between intervention groups showed significant variations before(P=0.028),during(P=0.047),and after(P=0.036)the intervention on day 7.Comparing GCS scores between the family members’voice stimulus group and the control group,significant changes were observed on days 5 and 7(P=0.043,0.030,0.030,and 0.014,0.012,0.012)before,during,and after the intervention.Conclusions:The use of beloved family members’voices proved more effective in elevating the patients’level of consciousness compared to both the nurse voice stimulus group and the control group.
文摘Although substantial research shows the effectiveness of written corrective feedback(WCF)in treating simple grammar structures,more research is still needed to refute Truscott’s claim that WCF may not work on complex grammar structures.Similarly,a previous body of research has shown that the degree of explicitness of feedback moderates the efficacy of WCF.However,most WCF studies have systematically manipulated only direct corrective feedback.The current study was therefore conducted to fill these gaps in the literature.To this end,five intact classes of Functional English were recruited and later randomly assigned to four treatment groups:DCF,DCF+ME,ICF,and ICF+ME,and one control group that received no feedback.All the groups took part in three WCF treatment sessions,during which they wrote two different pieces:a news report and a picture description.Later,only the treatment groups received the WCF.The WCF’s effectiveness was measured by writing tests and grammaticality judgment tasks(GJT).The results demonstrated that WCF helped L2 learners improve their grammatical accuracy of passive voice tenses.The study further showed that the group that received the most explicit type of WCF fared better than the ones that received the least explicit type of WCF.Important pedagogical implications for ESL/EFL teachers are discussed.
文摘From children to teenagers and then to adults,individuals’emotional expression ability has undergone significant changes.From the compulsory education stage to the general senior middle school stage,students’cognitive and emotional needs will change greatly.Students of different school stages show significant differences in their vocal emotional expression,which not only affects their social skills but also has important guiding significance for educational practice.Therefore,this study focuses on the vocal emotional expression of students at different school stages.This study explores the differences in vocal emotional expression among students at different school stages(primary school,junior high school,and senior high school),with a focus on the expression characteristics of different emotional types and the situation of inconsistent internal and external emotions.The research results show that in terms of emotional types,students at different school stages have a relatively high accuracy rate in expressing sadness,while the accuracy rate for expressing anger is the lowest,and the accuracy rate for expressing anger is significantly lower than that for other emotions.In terms of school stages,the overall accuracy of emotional expression among students improves with the increase of school stages,but the differences between school stages are not significant.In the task of expressing inconsistent internal and external emotions,the study found that senior high school students have significantly better emotional conversion ability than primary and junior high school students,indicating that as the school stage increases,students’adaptability in complex emotional expression tasks gradually improves.
基金Supported by the Icelandic Technological Development Fund(Project No.1910271).
文摘Background Vibrotactile feedback systems are widely used in assistive technology,wearable devices,and virtual environments to deliver precise tactile information.The timing of interstimulus intervals(ISIs)plays a critical role in determining how accurately users perceive and interpret vibrotactile patterns.The optimal use of ISIs can increase the effectiveness of these systems,improve user interaction,and enable reliable,intuitive feedback in diverse applications.We examined how different interstimulus intervals ISIs impact the accuracy of vibrotactile pattern recognition.Methods Participants wore a forearm-mounted device with six voice coil actuators arranged in a 3×2 grid,delivering Braille-based vibrotactile patterns sequentially at ISIs ranging from 10 to 2500 ms.Eight participants performed identification tasks involving Icelandic Braille patterns categorized as either short(2-3 actuators)or long(4-5 actuators).A repeated measures ANOVA was conducted to assess the effects of ISI,pattern type,and practice(across two testing blocks)on pattern recognition accuracy.Results For short patterns,accuracy was highest(92%-98%)at ISIs of 50-700 ms,with peak performance at 300 ms.For long patterns,accuracy reached 86%-94%at ISIs of 100-500 ms,peaking at 400 ms.Participants were more accurate with short patterns,and performance improved significantly over time for both short and long patterns,highlighting the importance of training for vibrotactile pattern recognition.Conclusions These results underscore the importance of careful selection of ISIs in vibrotactile feedback systems for accurate pattern identification.The findings provide valuable insights for conveying tactile information using wearable devices,contributing to better tactile feedback and performance in applications requiring precise vibrotactile information delivery.
基金funding from the Research Council of Lithuania(LMTLT),agreement No.S-MIP-23-46.
文摘This study introduces a novel voice cloning framework driven by Mordukhovich Subdifferential Optimization(MSO)to address the complex multi-objective challenges of pathological speech synthesis in underresourced Lithuanian language with unique phonemes not present in most pre-trained models.Unlike existing voice synthesis models that often optimize for a single objective or are restricted to major languages,our approach explicitly balances four competing criteria:speech naturalness,speaker similarity,computational efficiency,and adaptability to pathological voice patterns.We evaluate four model configurations combining Lithuanian and English encoders,synthesizers,and vocoders.The hybrid model(English encoder,Lithuanian synthesizer,English vocoder),optimized via MSO,achieved the highest Mean Opinion Score(MOS)of 4.3 and demonstrated superior intelligibility and speaker fidelity.The results confirm that MSO enables effective navigation of trade-offs in multilingual pathological voice cloning,offering a scalable path toward high-quality voice restoration in clinical speech applications.This work represents the first integration of Mordukhovich optimization into pathological TTS,setting a new benchmark for speech synthesis under clinical and linguistic constraints.
