Automatic Speech Recognition (ASR) engine to improve autistic children speech

自动语音识别（ASR）引擎可改善自闭症儿童的言语能力

• 批准号: 10056712
• 金额: $ 6.37万
• 依托单位: AUTSERA LTD
• 依托单位国家: 英国
• 项目类别: Grant for R&D
• 财政年份: 2023
• 资助国家: 英国
• 起止时间: 2023 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=10056712
• 关键词: Automatic; Speech; Recognition; ASR; engine

Social communication skills challenges affect over 400,000 children aged 3-11 in the UK, who are diagnosed with autism and other developmental speech delay challenges. Effective interventions and therapy exist and are most effective in early years. However, providing this crucial support is constrained by the scarcity of qualified professional resources and by the limitations of government funding.As a result, most children do not receive the effective and timely support they need. Research showed that this leads to deteriorating mental and physical health, social exclusion, poor quality of life and short lives.In this project, we combine opensource speech recognition technology and models with effective speech therapy methods, to build a prototype for an inclusive, innovative mobile app that helps autistic children and children with other speech delay challenges, improve their speech and communication skills.

社交技能挑战影响了英国超过40万名3-11岁的儿童，他们被诊断患有自闭症和其他发育性语言延迟挑战。有效的干预和治疗是存在的，并且在早年最有效。然而，由于缺乏合格的专业资源和政府资金有限，提供这一重要支助受到限制，因此，大多数儿童得不到他们所需的有效和及时的支助。研究表明，这会导致心理和身体健康恶化，社会排斥，生活质量差和寿命短。在这个项目中，我们将联合收割机开源语音识别技术和模型与有效的语音治疗方法相结合，构建一个包容性的创新移动的应用程序原型，帮助自闭症儿童和其他语言延迟挑战的儿童，提高他们的语言和沟通能力。