Is it possible to have a natural conversation with AI? The potential of speech recognition technology
Katsunobu Itou, Professor
Department of Digital Media, Faculty of Computer and Information Sciences
------
Speech recognition is a technology used in such applications as vehicle navigation and Siri. It analyzes the speech content of humans and turns it into text. Data processing for both language and sound are necessary for speech recognition. Prof. Itou is primarily engaged in research on statistical language models. The currently popular ChatGPT makes use of this technology. Prof. Itou has been conducting research in this field since the 1990s, and at that time he created a speech recognition toolkit that anyone can use. The tool has been used in many research projects across a long time. In his research, Prof. Itou emphasizes the idea of “doing what no one else can do, and being a pioneer.” And in recent times, he has expanded his field of expertise from “language” to “sound.” Speaking on the future, he says, “If data is accumulated and research advances, humans may be able to have natural conversation with AI, just like we’d have with our family or friends.”