[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-126568-105":59,"doc-detail-126568-en":130},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","a-review-of-speech-centric-trustworthy-machine-learning-privacy-safety-and-fairness","A Review of Speech-centric Trustworthy Machine Learning - Privacy, Safety, and Fairness","","Speech-centric machine learning systems have transformed transportation, healthcare, education, and defense by reshaping how people live, work, and communicate. Despite rapid progress, research has shown that many speech-centric ML systems face trust gaps in the form of privacy leakage, discriminatory or uneven performance, and susceptibility to adversarial attacks. To mitigate these risks, the work surveys privacy-, safety-, and fairness-related trustworthy approaches, providing a comprehensive synthesis for the research community and outlining promising future directions.",{"@graph":69,"@context":122},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/a-review-of-speech-centric-trustworthy-machine-learning-privacy-safety-and-fairness/126568/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/a-review-of-speech-centric-trustworthy-machine-learning-privacy-safety-and-fairness/126568.png","ImageObject",300,407,{"name":92,"@type":93},"Himbo","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-09-19","2026-08-05",true,{"@type":102,"interactionType":103,"userInteractionCount":39},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"What does the survey focus on in speech-centric trustworthy machine learning?","Question",{"text":112,"@type":113},"It focuses on privacy, safety, and fairness issues in speech-centric ML, and summarizes approaches used to address those risks.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"Why are trustworthy concerns important for speech-centric ML systems?",{"text":117,"@type":113},"Deployment across real-world applications creates challenges such as privacy breaches, discriminatory performance, and vulnerability to adversarial attacks.",{"name":119,"@type":110,"acceptedAnswer":120},"What future research directions are highlighted?",{"text":121,"@type":113},"The survey outlines upcoming work for privacy, safety, fairness, their trade-offs, and trustworthy multimodal applications.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},126568,1785933377,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":39,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":139,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":129,"read_time":144},687207017582,"https://ap-avatar.wpscdn.com/davatar_994ba38a5ba835b3df7d355c54d3ed8d","arXiv :2212 .09006v2 [ cs . SD] 16 Apr 2023  \nA Review of Speech-centric Trustworthy Machine Learning: Privacy, Safety, and Fairness  \nTiantian Feng, Rajat Hebbar†, Nicholas Mehlman†, Xuan Shi†, Aditya Kommineni† and Shrikanth Narayanan (2023),“A Review of Speech-centric Trustworthy Machine Learning: Privacy, Safety, and Fairness”. ,† : Contribute equally in terms of text.  \nThis article may be used only for the purpose of research, teaching, and/or private study. Commercial use or systematic downloading (by robots or other automatic processes) is prohibited without explicit Publisher approval.  \nContents  \n1 Introduction 2  \n2 Speech-centric Machine Learning 7  \n2. 1 Speech-centric ML tasks   7  \n2.2 Speech Datasets   9  \n2.3 Speech Modeling Approach ................. 10  \n3 Related Surveys in Trustworthy Speech-centric Machine Learning 12  \n3. 1 Privacy   12  \n3.2 Safety   13  \n3.3 Fairness . . . . . . . . . . . . . . . . . . . . . . . . . . . 13  \n4 Safety in Speech-centric Machine Learning 14  \n4. 1 Evasion Attacks and Defenses . . . . . . . . . . . . . . . . 14  \n4.2 Poisoning Attacks and Defenses   18  \n5 Privacy in Speech-centric Machine Learning 22  \n5.1 Taxonomies of Privacy-related Speech-centric ML ..... 22  \n5.2 Privacy Threats   23  \n5.3 Mitigation Mechanisms   27  \n5.4 Downstream Speech Applications   30  \n5.5 Training Paradigms   32  \n6 Bias and Fairness in Speech-centric Machine Learning 39  \n6. 1 Fairness in Machine Learning   39  \n6.2 Fairness in Speech-centric Applications   42  \n7 Future Directions 46  \n7. 1 Privacy   46  \n7.2 Safety   47  \n7.3 Fairness . . . . . . . . . . . . . . . . . . . . . . . . . . . 48  \n7.4 Balance between Fairness, Privacy, and Safety   48  \n7.5 Trustworthy Multimodal Applications   49  \n8 Conclusion 50  \n9 Acknowledgement 51  \nReferences 52  \nA Review of Speech-centric Trustworthy Machine Learning:  \nPrivacy, Safety, and Fairness  \nTiantian Feng 1 , Rajat Hebbar†1, Nicholas Mehlman†1, Xuan Shi†1, Aditya Kommineni†1 and Shrikanth Narayanan 1  \n1 University of Southern California, Los Angeles, USA; email: [tiantiaf@usc. edu](tiantiaf@usc. edu)  \nABSTRACT  \nSpeech-centric machine learning systems have revolutionized a number of leading industries ranging from transportation and healthcare to education and defense, fundamentally reshaping how people live, work, and interact with eachother. However, recent studies have demonstrated that many speech-centric ML systems may need to be considered more trustworthy for broader deployment. Specifically, concerns over privacy breaches, discriminating performance, and vulnerability to adversarial attacks have all been discovered in ML research fields. In order to address the above challenges and risks, a significant number of efforts have been made to ensure these ML systems are trustworthy, especially private, safe, and fair. In this paper, we conduct the first comprehensive survey on speech-centric trustworthy ML topics related to privacy, safety, and fairness. In addition to serving as a summary report for the research community, we highlight several promising future research directions to inspire researchers who wish to explore further in this area.  \n1  \nIntroduction  \nIn the last few years, machine learning (ML), particularly deep learning, has empowered tremendous breakthroughs in a variety of research fieldsand applications, including natural language processing (Devlin et al. , 2018), image classification (He et al. , 2016), video recommendation (Davidson et al. , 2010), healthcare analysis (Miotto et al. , 2018), and even mastering the chess game (Silver et al. , 2016) . The deep learning model typically consists of multiple processing layers with a combination of both linear and non-linear computations. Although training a deep learning model with the multi-layer architecture demands the accumulation of massive datasets and access to large-scale computational infrastructures (Bengio et al. , 2021), the trained model usually achieves state-o","cbCaikBrsDYGH2t5","https://ap.wps.com/l/cbCaikBrsDYGH2t5","pdf",1119701,77,"English","# Introduction\n# Speech-centric Machine Learning\n## Speech-centric ML tasks\n## Speech Datasets\n## Speech Modeling Approach\n# Related Surveys in Trustworthy Speech-centric Machine Learning\n## Privacy\n## Safety\n## Fairness\n# Safety in Speech-centric Machine Learning\n## Evasion Attacks and Defenses\n## Poisoning Attacks and Defenses\n# Privacy in Speech-centric Machine Learning\n## Taxonomies of Privacy-related Speech-centric ML\n## Privacy Threats\n## Mitigation Mechanisms\n## Downstream Speech Applications\n## Training Paradigms\n# Bias and Fairness in Speech-centric Machine Learning\n## Fairness in Machine Learning\n## Fairness in Speech-centric Applications\n# Future Directions\n## Privacy\n## Safety\n## Fairness\n## Balance between Fairness, Privacy, and Safety\n## Trustworthy Multimodal Applications\n# Conclusion\n# Acknowledgement\n# References","[{\"question\":\"What does the survey focus on in speech-centric trustworthy machine learning?\",\"answer\":\"It focuses on privacy, safety, and fairness issues in speech-centric ML, and summarizes approaches used to address those risks.\"},{\"question\":\"Why are trustworthy concerns important for speech-centric ML systems?\",\"answer\":\"Deployment across real-world applications creates challenges such as privacy breaches, discriminatory performance, and vulnerability to adversarial attacks.\"},{\"question\":\"What future research directions are highlighted?\",\"answer\":\"The survey outlines upcoming work for privacy, safety, fairness, their trade-offs, and trustworthy multimodal applications.\"}]","A Review of Speech-centric Trustworthy Machine Learning - Privacy, Safety, and Fairness | PDF",194]