The expression of emotions through speech is an essential element of human communication, yet individuals with speech impairments who rely on speech-generating devices (SGDs) to talk are largely deprived of this communicative channel. No one should be forced to sound blas when in fact they?re elated or to sound apathetic when they intend to scold. While certain expressive speech modes are offered by a few currently available SGDs, the range is extremely limited, and the speech often falls short of conveying the desired emotion or intention. Further, the few speech modes provided are generally restricted to a single voice. The proposed project aims to remedy this unfortunate situation by enhancing the rule-based Synfonica text-to- speech system (S-TTS) to make it easy and cost-efficient to add a broad range of consistently recognizable expressive speech modes, all automatically available for all voices offered by the system. S-TTS currently produces speech for American English in a neutral reading mode. In the project, five new expressive conversational modes will be added?neutral, happiness, sadness, and two types of anger (known as hot anger and cold anger). Each mode will be available for four voices: an already existing adult male voice, an already existing adult female voice, and two new child voices (one male and one female) to be developed in the project. Users of S-TTS will be able to customize these voices to create new ones through special voice customization options to be added to the system outside of the project. While the system will be tested through standard types of perceptual tests, the expressive modes will also be evaluated in real-life situations. For this purpose, a simple augmentative and alternative communication (AAC) app that incorporates the S-TTS voices will be developed, and individuals who use SGDs will be recruited to use this app in their daily lives. These individuals will be asked to provide feedback on the effectiveness of the expressive modes for their interactions with others. In general, the project will be a major step toward enabling SGD users to speak in the manner they wish. They will at long last be empowered to express themselves through voices that can reflect their feelings and help them realize their intentions.
Currently, individuals with speech impairments who rely on speech-generating devices for their verbal communication are deprived of the ability to realistically express their emotions through speech. The proposed project aims to address this problem by enhancing Synfonica?s text-to-speech system with expressive speech capabilities. With such an enhanced system, developers of speech-generating devices will be able to provide users of these devices with genuinely expressive voices.
