Millions of people reach out to digital assistants such as Siri every day, asking for information, making phone calls, seeking assistance, and much more. The expectation is that such assistants should understand the intent of the user's query. Detecting the intent of a query from a short, isolated utterance is a difficult task. Intent cannot always be obtained from speechrecognized transcriptions. A transcription-driven approach can interpret what has been said but fails to acknowledge how it has been said, and as a consequence, may ignore the expression present in the voice. Our work investigates whether a system can reliably detect vocal expression in queries using acoustic and paralinguistic embedding. Results show that the proposed method offers a relative equal error rate (EER) decrease of 60% compared to a bag-of-word based system, corroborating that expression is significantly represented by vocal attributes, rather than being purely lexical. Addition of emotion embedding helped to reduce the EER by 30% relative to the acoustic embedding, demonstrating the relevance of emotion in expressive voice.

No abstract

While the needs and applications for computing technology have been growing, the enrollment and interest in Computer Science (CS) at the university level has not been growing in proportion. Moreover, the increasing prevalence of globalization requires a new set of skills for future technology leaders. These skills include the ability to work well in multidisciplinary and globally distributed teams, cross cultural boundaries with ease, create innovative solutions for problems that arise in unfamiliar settings, and think outside the box to solve a variety of problems while building effectively upon related work in the literature. These changes are also reflected in the growing interest among youth in global issues and how we can make a difference in underserved communities. This paper reports on our experience in designing and deploying an innovative internship that addresses these issues and seeks to enhance participation and education in CS through guided research projects in underserved communities.

Abstract-Although para-social workers carry the primary responsibility in providing essential services to the growing population of orphans and vulnerable children in Tanzania, they are often not paid for this work. Moreover, these para-social workers are unable to access governmental resources due to the lack of an efficient means of reporting their needs to relevant government officials in a timely manner. In this paper we describe a text message (SMS) based solution that harnesses the prevalence of mobile phones coupled with several Open Source tools to empower these para-social workers. Specifically, we build a more efficient mechanism for reporting summary data on orphans and vulnerable children to relevant government officials in a cost-effective and efficient manner. This paper reports on our needs assessment process, reviews the related work, describes the implementation and testing of our prototype solution, and concludes with a discussion of relevant future work.

Despite the importance of literacy to most aspects of life, underserved communities continue to suffer from low literacy rates; especially for globally prevalent languages such as English. This is particularly problematic since English literacy is an important skill for increased access to income generation and higher education opportunities in these communities. Teaching English in underserved communities can be challenging due to a variety of obstacles. In this paper we report on our work with a primary school in Tanzania, with migrant workers in Doha, and with a school for the deaf in the USA to address some of these challenges using web technology and mobile phones. We developed a web-accessible content authoring tool that teachers can use to create and modify English grammar exercises, and a simple game that students use on a mobile phone to practice the content uploaded from the authoring tool. Our solution is primarily intended to increase the motivation for practicing English grammar among students, and thereby enhance English literacy in underserved communities. Feedback from teachers and students in all of our field studies indicate this solution is affordable and has significant potential to enhance the English classroom and homework experience.