Technology-based tools and features like text-to-speech (TTS), voice-to-speech, and object recognition can help people with partially visual and speech impairments access a variety of features, such as voice-based email, virtual navigation, and object recognition.
Author
Harshit Garg, Department of Computer Science & Engineering, Bharati Vidyapeeth’s College of Engineering, New Delhi, Delhi, India
Srishti Jhunthra, Department of Computer Science & Engineering, Bharati Vidyapeeth’s College of Engineering, New Delhi, Delhi, India
Madhav Kindra, Department of Computer Science & Engineering, Bharati Vidyapeeth’s College of Engineering, New Delhi, Delhi, India
Vikrant Dixit, Department of Computer Science & Engineering, Bharati Vidyapeeth’s College of Engineering, New Delhi, Delhi, India
Vedika Gupta, Associate Professor, Jindal Global Business School, O.P. Jindal Global University, Sonipat, Haryana, India
Summary
With the modern advancements in technology, every individual nowadays is moving over to an easier and more effective lifestyle. People are moving on with technology and finding solutions to problems faced in everyday life. Normal people are getting privileges of technology but sometimes the benefits could not reach the partially disabled ones.
Partially disabled people face several problems in their day-to-day lives from navigation to communicating with others. The partially sighted people and hearing-impaired people try to cope with the normal ones but they do not get many opportunities. This chapter focuses on partially disabled people to provide them with some of the features to overcome a few of the problems faced in the real world.
This chapter demonstrates the aid for the partially visual and hearing impaired through communication via voice for the visually impaired and communication via text for the hearing impaired. This chapter is divided into two parts, initially consisting of text-to-speech (TTS) and voice-to-speech capabilities, and object recognition for people with disabilities.
This chapter includes a brief analysis of various models and algorithms such as interactive speech response, convolutional neural network, recurrent neural network, and TTS. Another part is the integration with Android applications. Here, the trained deep learning model serves as the source for the backend in object detection. Models are imported to predict outcomes, and TTS helps people with disabilities to access a variety of features, such as voice-based email, object recognition, and virtual navigation.
Published in: Uncertainty in Computational Intelligence-Based Decision Making
To read the full article, please click here.
