Using advanced AI and Machine Learning to identify and detect signs of vulnerability, wellbeing and mental stress over the phone

使用先进的人工智能和机器学习通过电话识别和检测脆弱性、健康和精神压力的迹象

• 批准号: 47656
• 金额: $ 34.54万
• 依托单位: VOICEIQ LIMITED
• 依托单位国家: 英国
• 项目类别: Study
• 财政年份: 2020
• 资助国家: 英国
• 起止时间: 2020 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=47656
• 关键词: Using; advanced; AI; Machine; Learning

By combining leading research in psychology and linguistics (developed as part of a KTP with Manchester University) with advanced Natural Language Processing (NLP), waveform analysis and machine learning in our platform, we intend to develop a highly advanced , highly accurate vulnerability analysis tool that will help Contact Centres organisations better serve their customers, prevent non- compliance with vulnerability legislation and help monitor the mental wellbeing of call operatives.Although calls are recorded as a matter of course in all contact centres, at present there is no automated system to detect vulnerability, general wellbeing using voice data, with on average only a 2% sample analysed manually. By developing a system that can detect the signs of vulnerability both in the tone of voice and in the language used, the innovative technology, will provide the following positive impacts;**Client benefits*** Improved accuracy in detecting stress and vulnerability.* Provide an automated, cost effective solution to an expensive, manual process.* Monitor 100% of calls rather than <2% call presently.* Reduce the risk of fines & penalties for non-compliance* Intelligent customer service process - building in emotional intelligence into the relationship.**Consumer benefits*** Prevent misselling/pressured sales to vulnerable people* Improved customer service/satisfaction.* Reduced incidents of misselling/cancelled orders during cool-off/fines.* Compliance with FCA Consumer Vulnerability legislation.* 90% saving (time and cost) assessing vulnerability.We see this as a £47m opportunity by 2026 and chance to develop a cutting- edge, best in class technology, with genuine global appeal - addressing a clear, global business need (inefficient manual processes), creating a number of high-value UK R&D jobs, boosting UK exports and helping companies address a defined regulatory requirement they have so far struggled with.Although initially focused on financial services and call centres, with significant wider technology potential in the mental healthcare sector once fully developed, the project represents a step-change innovation and a opportunity for the UK to take the lead in the development of mental health technology.

通过结合心理学和语言学的领先研究（作为与曼彻斯特大学KTP的一部分开发）在我们的平台中使用先进的自然语言处理（NLP），波形分析和机器学习，我们打算开发一个高度先进，高度准确的漏洞分析工具，帮助联络中心组织更好地为客户服务，防止不遵守脆弱性立法，并帮助监测呼叫操作员的精神健康。尽管所有呼叫中心都理所当然地记录了呼叫，但目前还没有自动化系统来检测脆弱性，使用语音数据的一般健康状况，平均只有2%的样品被人工分析。通过开发一种能够从语调和所用语言中发现脆弱性迹象的系统，这项创新技术将产生以下积极影响：** 客户受益 * 提高发现压力和脆弱性的准确性。为昂贵的手动流程提供自动化、经济高效的解决方案。*监控100%的呼叫，而不是目前的<2%呼叫。降低违规罚款和处罚的风险 * 智能客户服务流程-在关系中建立情商。**消费者利益 * 防止向弱势群体进行错误销售/压力销售 * 改善客户服务/满意度。*减少在冷静期/罚款期间的不当销售/取消订单事件。*遵守FCA消费者漏洞法规。*节省90%（时间和成本）评估脆弱性。我们认为这是一个到2026年价值4. 7亿英镑的机会，也是开发尖端、一流技术的机会，具有真正的全球吸引力--满足明确的全球业务需求（低效的手工流程），创造了大量高价值的英国研发工作，促进英国出口，帮助企业满足迄今为止一直难以满足的明确监管要求。尽管最初的重点是金融服务和呼叫中心，一旦充分开发，精神卫生保健领域的技术潜力将显著扩大，该项目代表了一个渐进式的创新，也是英国在精神卫生技术发展方面处于领先地位的机会。